UAIX.LmRuntime API Reference

Initializes a new LocalGgufGeneratedToken instance with validated dependencies and operational bounds.

sequence

The zero-based selection sequence within the current generation operation.

tokenId

The token identifier to process; it must fall within the validated vocabulary and operation-specific range.

selectedLogit

The deterministic argmax logit associated with the selected token. NaN is rejected; positive and negative infinity are retained because the lower-level greedy sampler compares infinities deterministically and the runtime verifies the exact observed value against the completed generation result.

Initializes a new LocalGgufVerificationException instance with validated dependencies and operational bounds.

message

The display-safe diagnostic message describing the failure without embedding prompt text, generated text, credentials, or private file contents.

expectedSha256

The expected lowercase or uppercase 64-character SHA-256 digest used to authenticate the complete artifact; a mismatch fails the operation before a trusted result is published.

actualSha256

The observed SHA-256 digest retained as path-free diagnostic evidence when available, or null when hashing did not complete.

innerException

The optional lower-level failure used only to select a bounded, path-free diagnostic category. The original exception object, message, stack trace, data, and file-name properties are never retained because they can contain private model paths or other caller-owned content.

expectedByteCount

The required file length when byte-count evidence is supplied.

actualByteCount

The actual byte count used to bound this operation; it must be nonnegative and within the supported range.

Tokenizes text using exact metadata loaded from the verified GGUF artifact.

text

The text processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

options

The optional TokenizationOptions controlling Tokenize; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The exact integer token identifiers and optional trace.

Decodes token identifiers using the tokenizer metadata from the verified GGUF artifact.

tokenIds

The token identifiers to decode in sequence order.

options

The optional MetadataDrivenGgufTokenizerDetokenizationOptions controlling Decode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Rejects creation of a LocalEndpoint inference session when required UAIX load-session context is absent.

Returns: No value is returned; this compatibility member always throws to require an explicit session-scoped UAIX context.

Creates a new inference session with independent sequence state and validated UAIX load-session evidence.

sessionContext

The LocalEndpoint worker session identifier and display-safe UAIX context.

Returns: An isolated deterministic session owned by the caller.

Releases the memory-mapped model after all child sessions have been disposed.

Initializes an unscoped model facade for compatibility with callers that do not require application labels.

Initializes a runtime facade for one LocalEndpoint application instance and optional legacy package label.

scope

The immutable runtime scope whose name and instance identifier are validated once and then retained as path-free correlation evidence for model verification operations.

Validates LocalEndpoint-supplied uaixRuntimeContext metadata without opening package or wiki files.

context

The display-safe profile, load-session, and long-term memory routing metadata.

Returns: An immutable path-relative evidence projection with a canonical context SHA-256.

Verifies the current bytes and file-policy boundaries of one local GGUF artifact without loading a model.

modelPath

The local file-system model path processed by this operation; it must satisfy the containing component's path and scope policy.

expectedSha256

The expected lowercase or uppercase 64-character SHA-256 digest used to authenticate the complete artifact; a mismatch fails the operation before a trusted result is published.

options

The optional LocalGgufModelLoadOptions controlling VerifyLocalModelFile; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: A path-free identity containing the verified digest and current byte count.

Verifies LocalEndpoint-reviewed associated artifacts without exposing private local paths.

associatedArtifacts

The LocalEndpoint-reviewed associated artifact load inputs to verify before model publication or worker execution proceeds.

options

Optional bounded parser, binding, containment, and file-validation controls shared with primary model verification.

Returns: Path-free associated-artifact identities sorted by role, file name, and SHA-256 digest.

Verifies current local model bytes against caller-owned digest and byte-count evidence without loading a model.

modelPath

The local file-system model path processed by this operation; it must satisfy the containing component's path and scope policy.

expectation

The path-free digest and byte-count evidence required by LocalEndpoint.

options

The optional LocalGgufModelLoadOptions controlling VerifyLocalModelFile; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: A path-free identity containing the verified digest and current byte count.

Loads a local GGUF model only after its exact SHA-256 digest and file-policy constraints are verified.

modelPath

The local file-system model path processed by this operation; it must satisfy the containing component's path and scope policy.

expectedSha256

The expected lowercase or uppercase 64-character SHA-256 digest used to authenticate the complete artifact; a mismatch fails the operation before a trusted result is published.

options

Optional bounded parser, binding, containment, and file-validation controls.

Returns: An owned mapped model that must be disposed after all child sessions are disposed.

Loads a local GGUF model after binding current file bytes to caller-owned digest and byte-count evidence.

modelPath

The local file-system model path processed by this operation; it must satisfy the containing component's path and scope policy.

expectation

The path-free digest and byte-count evidence required by LocalEndpoint.

options

Optional bounded parser, binding, containment, and file-validation controls.

Returns: An owned mapped model that must be disposed after all child sessions are disposed.

Loads a local GGUF model after binding the primary artifact and associated artifacts to caller-owned evidence.

modelPath

The local file-system model path processed by this operation; it must satisfy the containing component's path and scope policy.

expectation

The path-free digest and byte-count evidence required by LocalEndpoint for the primary GGUF artifact.

associatedArtifacts

The LocalEndpoint-reviewed associated artifacts to verify before the model is published.

options

Optional bounded parser, binding, containment, and file-validation controls.

Returns: An owned mapped model that must be disposed after all child sessions are disposed.

Runs bounded deterministic greedy generation and returns exact token identifiers with decoded text.

request

The LocalGgufGenerationRequest containing the complete caller-owned inputs for GenerateGreedy; required fields are validated and mutable collections are snapshotted before state changes or large allocations.

cancellationToken

A token observed before work and between committed model steps.

Returns: The generated token identifiers, text, stop reason, position, and final selected logit.

Runs bounded deterministic greedy generation and synchronously reports each selected token.

request

The LocalGgufGenerationRequest containing the complete caller-owned inputs for GenerateGreedy; required fields are validated and mutable collections are snapshotted before state changes or large allocations.

tokenObserver

The caller-owned observer invoked once for each selected token.

cancellationToken

A token observed before work and between committed model steps.

Returns: The generated token identifiers, text, stop reason, position, and final selected logit.

Clears sequence position and key/value-cache state retained by this session.

Releases isolated session state without disposing the shared model owner.

Serializes complete session state and appends a SHA-256 digest.

snapshot

The immutable state snapshot being validated, serialized, restored, or analyzed without retaining caller-owned mutable aliases.

options

The optional LlamaSessionPersistenceOptions controlling Serialize; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: A newly allocated byte[] containing the ordered result of LlamaSessionArtifactSerializer.Serialize: Serializes complete session state and appends a SHA-256 digest. The caller owns the returned array and later mutation cannot alter the source object.

Deserializes the llama session artifact from the validated persisted representation.

bytes

The bytes sequence used by this operation; its required length, ordering, and element bounds are validated before access.

maximumByteCount

The maximum byte count used to bound this operation; it must be nonnegative and within the supported range.

Returns: The LlamaSessionArtifact result produced by LlamaSessionArtifactSerializer.Deserialize for this contract: Deserializes the llama session artifact from the validated persisted representation. It is published only after all documented validation and ownership transitions succeed.

Writes a complete artifact to a local file.

path

The local file-system path processed by this operation; it must satisfy the containing component's path and scope policy.

snapshot

The immutable state snapshot being validated, serialized, restored, or analyzed without retaining caller-owned mutable aliases.

options

The optional LlamaSessionPersistenceOptions controlling Save; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The LlamaSessionArtifact result produced by LlamaSessionArtifactSerializer.Save for this contract: Writes a complete artifact to a local file. It is published only after all documented validation and ownership transitions succeed.

Reads and verifies a complete artifact from a local file.

path

The local file-system path processed by this operation; it must satisfy the containing component's path and scope policy.

maximumByteCount

The maximum byte count used to bound this operation; it must be nonnegative and within the supported range.

Returns: The verified artifact, with ownership and disposal obligations defined by the returned type and the Load contract.

Verifies the supplied fixture directory and returns bounded evidence only after every required check succeeds.

fixtureDirectory

The directory containing a fixture manifest and GGUF artifact.

Returns: The FixtureVerificationResult result produced by FixtureDirectoryVerifier.Verify for this contract: Verifies the supplied fixture directory and returns bounded evidence only after every required check succeeds. It is published only after all documented validation and ownership transitions succeed.

Initializes a mapped LLaMA weight set from a complete binding manifest.

mapping

The mapped GGUF file that owns tensor storage.

manifest

The validated manifest that binds tensor requirements, model identity, and storage diagnostics used by the operation.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

Retrieves the llama bound tensor from the current LlamaBoundWeightSet state after validating the requested access.

role

The semantic LLaMA tensor role used to select the required bound tensor from the validated manifest.

blockIndex

The zero-based block index; it must identify an existing position within the relevant validated range.

Returns: The LlamaBoundTensor result produced by LlamaBoundWeightSet.Get for this contract: Retrieves the llama bound tensor from the current LlamaBoundWeightSet state after validating the requested access. It is published only after all documented validation and ownership transitions succeed.

Materializes bounded float32 arrays for the scalar correctness runtime.

maximumCopiedBytes

The maximum total bytes that may be copied from mapped storage.

Returns: The immutable reference weights and explicit copy ledger.

Loads and validates one mapped local model.

path

The local file-system path processed by this operation; it must satisfy the containing component's path and scope policy.

options

The optional LlamaMappedModelLoadOptions controlling Load; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The owned mapped model, with ownership and disposal obligations defined by the returned type and the Load contract.

Creates an independent scalar reference session with its own key/value state.

Returns: The new mapped reference session, with ownership and disposal obligations defined by the returned type and the CreateReferenceSession contract.

Creates an independent compatibility session over explicitly materialized float32 arrays.

Returns: The materialized compatibility session, with ownership and disposal obligations defined by the returned type and the CreateMaterializedReferenceSession contract.

Gets the bounded reference materialization evidence, creating it on first use.

Returns: The LlamaReferenceWeightMaterialization result produced by LlamaMappedModel.GetReferenceMaterialization for this contract: Gets the bounded reference materialization evidence, creating it on first use. It is published only after all documented validation and ownership transitions succeed.

Releases resources owned by LlamaMappedModel and transitions it to the disposed state.

Resets this session's sequence and key/value state.

Tokenizes a prompt, executes prefill, selects argmax, and decodes exactly one token.

prompt

The prompt processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

options

The optional LlamaOneTokenOptions controlling DecodeOneGreedy; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The LlamaMappedGreedyTokenResult result produced by LlamaMappedReferenceSession.DecodeOneGreedy for this contract: Tokenizes a prompt, executes prefill, selects argmax, and decodes exactly one token. It is published only after all documented validation and ownership transitions succeed.

Tokenizes a prompt and generates greedy token identifiers into caller-owned buffers.

prompt

The prompt processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

generatedTokenIds

The ordered token identifiers to process; sequence order is preserved and each identifier is validated where required.

finalLogits

The final logits sequence used by this operation; its required length, ordering, and element bounds are validated before access.

generationOptions

The generation options that define validation limits and execution behavior; required values are checked before use.

tokenizationOptions

The tokenization options that define validation limits and execution behavior; required values are checked before use.

cancellationToken

The caller-provided token used to cancel the operation before additional work or results are published.

Returns: The LlamaGreedyGenerationResult result produced by LlamaMappedReferenceSession.GenerateGreedy for this contract: Tokenizes a prompt and generates greedy token identifiers into caller-owned buffers. It is published only after all documented validation and ownership transitions succeed.

Tokenizes a prompt, generates greedy token identifiers, and reports each selected token synchronously.

prompt

The prompt processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

generatedTokenIds

The ordered token identifiers to process; sequence order is preserved and each identifier is validated where required.

finalLogits

The final logits sequence used by this operation; its required length, ordering, and element bounds are validated before access.

generationOptions

The generation options that define validation limits and execution behavior; required values are checked before use.

tokenizationOptions

The tokenization options that define validation limits and execution behavior; required values are checked before use.

tokenObserver

The optional observer invoked once for each selected token.

cancellationToken

A token observed before selection and before the next committed model step.

Returns: The LlamaGreedyGenerationResult result produced by LlamaMappedReferenceSession.GenerateGreedy for this contract: Tokenizes a prompt, generates greedy token identifiers, and reports each selected token synchronously. It is published only after all documented validation and ownership transitions succeed.

Exports complete deterministic state with model, configuration, tokenizer, and cache-layout identities.

options

Optional persistence metadata. Empty identity fields are resolved from the mapped model.

Returns: A newly allocated byte[] containing the ordered result of LlamaMappedReferenceSession.ExportState: Exports complete deterministic state with model, configuration, tokenizer, and cache-layout identities. The caller owns the returned array and later mutation cannot alter the source object.

Saves complete deterministic state to a local artifact.

path

The local file-system path processed by this operation; it must satisfy the containing component's path and scope policy.

options

The optional LlamaSessionPersistenceOptions controlling SaveState; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The LlamaSessionArtifact result produced by LlamaMappedReferenceSession.SaveState for this contract: Saves complete deterministic state to a local artifact. It is published only after all documented validation and ownership transitions succeed.

Restores verified complete state after enforcing mapped model and tokenizer identities.

bytes

The bytes sequence used by this operation; its required length, ordering, and element bounds are validated before access.

maximumByteCount

The maximum byte count used to bound this operation; it must be nonnegative and within the supported range.

Returns: The LlamaSessionArtifact result produced by LlamaMappedReferenceSession.RestoreState for this contract: Restores verified complete state after enforcing mapped model and tokenizer identities. It is published only after all documented validation and ownership transitions succeed.

Loads and restores complete deterministic state from a local artifact.

path

The local file-system path processed by this operation; it must satisfy the containing component's path and scope policy.

maximumByteCount

The maximum byte count used to bound this operation; it must be nonnegative and within the supported range.

Returns: The verified artifact, with ownership and disposal obligations defined by the returned type and the LoadState contract.

Releases resources owned by LlamaMappedReferenceSession and transitions it to the disposed state.

Creates a LLaMA-family configuration from GGUF metadata.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

Returns: The LlamaModelConfig result produced by LlamaModelConfig.FromGguf for this contract: Creates a LLaMA-family configuration from GGUF metadata. It is published only after all documented validation and ownership transitions succeed.

Validates architectural invariants required by the scalar LLaMA runtime.

Applies the LLaMA RMSNorm operation through the CPU reference kernel.

input

The source data consumed by the operation; caller-owned storage is not retained after the method returns.

weight

The weight sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

epsilon

The positive normalization epsilon added to the mean-square term to avoid division by zero while preserving deterministic numerical behavior.

Applies LLaMA-style RoPE to a query or key vector in place.

vector

The vector sequence used by this operation; its required length, ordering, and element bounds are validated before access.

cos

The cos sequence used by this operation; its required length, ordering, and element bounds are validated before access.

sin

The sin sequence used by this operation; its required length, ordering, and element bounds are validated before access.

ropeDimensions

The even number of leading head dimensions transformed by rotary positional encoding.

Validates all reference-weight shapes against a LLaMA configuration.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

Initializes a reference session through the v1.8.0 array-backed compatibility path.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

weights

The validated model-weight source or bound weight set consumed read-only by the deterministic reference operation.

tokenizer

The optional tokenizer used only to decode the selected token text.

Initializes a reference session over immutable array-backed or direct mapped weight sources.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

weights

The validated model-weight source or bound weight set consumed read-only by the deterministic reference operation.

tokenizer

The optional tokenizer used only to decode the selected token text.

Clears sequence state and all key/value cache contents.

Captures complete deterministic session state without serializing live model pointers.

Returns: The LlamaReferenceSessionSnapshot result produced by LlamaReferenceSession.CaptureState for this contract: Captures complete deterministic session state without serializing live model pointers. It is published only after all documented validation and ownership transitions succeed.

Restores complete deterministic state after validating sequence, vocabulary, and cache identities.

snapshot

The immutable state snapshot being validated, serialized, restored, or analyzed without retaining caller-owned mutable aliases.

Evaluates one input token and writes next-token logits.

tokenId

The token identifier to process; it must fall within the validated vocabulary and operation-specific range.

logits

The destination buffer with at least vocabulary-size elements.

Evaluates a prompt and returns exactly one greedily selected next token.

promptTokenIds

The ordered token identifiers to process; sequence order is preserved and each identifier is validated where required.

resetSession

Whether existing key/value state should be cleared first.

Returns: The LlamaGreedyTokenResult result produced by LlamaReferenceSession.DecodeOneGreedy for this contract: Evaluates a prompt and returns exactly one greedily selected next token. It is published only after all documented validation and ownership transitions succeed.

Evaluates every prompt token and leaves the final logits available for deterministic selection.

promptTokenIds

The ordered token identifiers to process; sequence order is preserved and each identifier is validated where required.

resetSession

Whether existing sequence and cache state should be cleared first.

Copies the most recently computed logits to a caller-provided destination.

destination

The destination with room for the configured vocabulary.

Generates deterministic greedy token identifiers into caller-owned buffers.

promptTokenIds

The ordered token identifiers to process; sequence order is preserved and each identifier is validated where required.

generatedTokenIds

The ordered token identifiers to process; sequence order is preserved and each identifier is validated where required.

finalLogits

The caller-owned destination for the final available logits.

options

The optional LlamaGreedyGenerationOptions controlling GenerateGreedy; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

cancellationToken

The caller-provided token used to cancel the operation before additional work or results are published.

Returns: The LlamaGreedyGenerationResult result produced by LlamaReferenceSession.GenerateGreedy for this contract: Generates deterministic greedy token identifiers into caller-owned buffers. It is published only after all documented validation and ownership transitions succeed.

Generates deterministic greedy token identifiers and reports each selection to a synchronous observer.

promptTokenIds

The ordered token identifiers to process; sequence order is preserved and each identifier is validated where required.

generatedTokenIds

The ordered token identifiers to process; sequence order is preserved and each identifier is validated where required.

finalLogits

The caller-owned destination for the final available logits.

options

The optional LlamaGreedyGenerationOptions controlling GenerateGreedy; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

tokenObserver

The optional synchronous observer invoked once after each token is selected.

cancellationToken

A token observed before selection and before the next committed model step.

Returns: The LlamaGreedyGenerationResult result produced by LlamaReferenceSession.GenerateGreedy for this contract: Generates deterministic greedy token identifiers and reports each selection to a synchronous observer. It is published only after all documented validation and ownership transitions succeed.

Selects and decodes one greedy token from the current logits.

promptTokenCount

The number of prompt tokens associated with the current logits.

Returns: The LlamaGreedyTokenResult result produced by LlamaReferenceSession.SelectGreedyToken for this contract: Selects and decodes one greedy token from the current logits. It is published only after all documented validation and ownership transitions succeed.

Creates the session from the validated inputs required by LlamaReferenceFixture.

Returns: A session with empty key/value cache state.

Creates a one-block, five-token deterministic LLaMA fixture.

Returns: The fixture configuration, weights, tokenizer, and prompt, with ownership and disposal obligations defined by the returned type and the CreateDeterministic contract.

Decodes the next token for the active sequence.

tokenId

The token identifier to process; it must fall within the validated vocabulary and operation-specific range.

cancellationToken

The caller-provided token used to cancel the operation before additional work or results are published.

Returns: An asynchronous ValueTask<int> that completes with the result of ILlamaSession.DecodeAsync: Decodes the next token for the active sequence. Fault and cancellation states are propagated without a successful partial result.

Executes a minimal reference forward pass over hidden-state logits.

hiddenState

The hidden state sequence used by this operation; its required length, ordering, and element bounds are validated before access.

weights

The LLaMA weights used by the reference path.

logits

The logits sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Applies the supplied input to the supplied values while preserving the operation's numeric and shape invariants.

input

The source data consumed by the operation; caller-owned storage is not retained after the method returns.

weights

The weights sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

epsilon

The positive normalization epsilon added to the mean-square term to avoid division by zero while preserving deterministic numerical behavior.

Applies rotary position embedding to adjacent hidden-state pairs.

values

The values sequence used by this operation; its required length, ordering, and element bounds are validated before access.

position

The zero-based sequence or cache position addressed by the operation; it must lie within the allocated context and readable or writable range.

theta

The rotary angle in radians applied to the paired vector components at the addressed position.

Applies a minimal causal attention score computation.

query

The query sequence used by this operation; its required length, ordering, and element bounds are validated before access.

keys

The keys sequence used by this operation; its required length, ordering, and element bounds are validated before access.

values

The values sequence used by this operation; its required length, ordering, and element bounds are validated before access.

headSize

The numeric head size consumed by ApplyCausal; it must satisfy the member's documented range, geometry, and finite-value requirements.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Maps an attention query head to the corresponding KV head.

queryHead

The zero-based query-head index mapped deterministically to its corresponding key/value head.

queryHeadCount

The query head count used to bound this operation; it must be nonnegative and within the supported range.

keyValueHeadCount

The key value head count used to bound this operation; it must be nonnegative and within the supported range.

Returns: The int value computed by GroupedQueryAttentionMap.MapHead for this contract: Maps an attention query head to the corresponding KV head. Range, finite-value, and overflow checks are completed before the value is returned.

Applies the SwiGLU activation to validated gate and up-projection vectors.

gate

The gate sequence used by this operation; its required length, ordering, and element bounds are validated before access.

up

The up sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Computes logits from a hidden vector and a row-major projection matrix.

hiddenState

The hidden state sequence used by this operation; its required length, ordering, and element bounds are validated before access.

projection

The projection sequence used by this operation; its required length, ordering, and element bounds are validated before access.

logits

The logits sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Validates the absolute and relative parity tolerances used for numerical comparison.

Compares two logit vectors using absolute-or-relative error acceptance.

reference

The reference sequence used by this operation; its required length, ordering, and element bounds are validated before access.

candidate

The candidate sequence used by this operation; its required length, ordering, and element bounds are validated before access.

tolerance

The tolerance input of type LlamaParityTolerance read by LlamaLogitComparator.Compare; it must satisfy the member-specific nullability, identity, range, and ownership rules before dependent work begins.

Returns: The LlamaLogitComparison result produced by LlamaLogitComparator.Compare for this contract: Compares two logit vectors using absolute-or-relative error acceptance. It is published only after all documented validation and ownership transitions succeed.

Runs one reference model and one or more candidate models with identical prompt settings.

referenceModelPath

The local file-system reference model path processed by this operation; it must satisfy the containing component's path and scope policy.

candidateModelPaths

The local file-system candidate model paths processed by this operation; it must satisfy the containing component's path and scope policy.

prompt

The prompt processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

tolerance

The tolerance input of type LlamaParityTolerance read by LlamaStorageParityRunner.Run; it must satisfy the member-specific nullability, identity, range, and ownership rules before dependent work begins.

Returns: The LlamaStorageParityResult result produced by LlamaStorageParityRunner.Run for this contract: Runs one reference model and one or more candidate models with identical prompt settings. It is published only after all documented validation and ownership transitions succeed.

Attempts to find one bound tensor by semantic role and optional block index.

role

The semantic LLaMA tensor role used to select the required bound tensor from the validated manifest.

blockIndex

The zero-based block index; it must identify an existing position within the relevant validated range.

entry

When the method returns, contains the entry produced by the operation when successful; otherwise contains the type's default value.

Returns: True when try get binding succeeds for the supplied values; otherwise, false.

Throws when the manifest contains one or more diagnostics.

Initializes a binding exception from a failed manifest.

manifest

The validated manifest that binds tensor requirements, model identity, and storage diagnostics used by the operation.

Creates the required tensor list for the configuration.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

Returns: An ordered read-only IReadOnlyList<LlamaTensorRequirement> result from LlamaRequiredTensorRegistry.Build: Creates the required tensor list for the configuration. Mutable internal collection aliases are not exposed through the returned contract.

Binds required tensors from a parsed GGUF artifact using default validation options.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

Returns: The TensorBindingManifest result produced by LlamaTensorBinder.Bind for this contract: Binds required tensors from a parsed GGUF artifact using default validation options. It is published only after all documented validation and ownership transitions succeed.

Binds required tensors from a parsed GGUF artifact.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

options

The optional TensorBindingOptions controlling Bind; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The TensorBindingManifest result produced by LlamaTensorBinder.Bind for this contract: Binds required tensors from a parsed GGUF artifact. It is published only after all documented validation and ownership transitions succeed.

Initializes a new MappedFloat16VectorSource instance with validated dependencies and operational bounds.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Copies the to into caller-owned storage after validating the requested range and capacity.

destination

The destination buffer that receives the produced values.

Initializes a new MappedBFloat16VectorSource instance with validated dependencies and operational bounds.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Copies the to into caller-owned storage after validating the requested range and capacity.

destination

The destination buffer that receives the produced values.

Initializes a new MappedFloat16MatrixSource instance with validated dependencies and operational bounds.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Copies the row to into caller-owned storage after validating the requested range and capacity.

rowIndex

The zero-based row index; it must identify an existing position within the relevant validated range.

destination

The destination buffer that receives the produced values.

Multiplies the supplied vector by the supplied vector without changing logical row order.

vector

The vector sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Initializes a new MappedBFloat16MatrixSource instance with validated dependencies and operational bounds.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Copies the row to into caller-owned storage after validating the requested range and capacity.

rowIndex

The zero-based row index; it must identify an existing position within the relevant validated range.

destination

The destination buffer that receives the produced values.

Multiplies the supplied vector by the supplied vector without changing logical row order.

vector

The vector sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Initializes a new MappedQ4_KMatrixSource instance with validated dependencies and operational bounds.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Copies the row to into caller-owned storage after validating the requested range and capacity.

rowIndex

The zero-based row index; it must identify an existing position within the relevant validated range.

destination

The destination buffer that receives the produced values.

Multiplies the supplied vector by the supplied vector without changing logical row order.

vector

The vector sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Initializes a new MappedQ6_KMatrixSource instance with validated dependencies and operational bounds.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Copies the row to into caller-owned storage after validating the requested range and capacity.

rowIndex

The zero-based row index; it must identify an existing position within the relevant validated range.

destination

The destination buffer that receives the produced values.

Multiplies the supplied vector by the supplied vector without changing logical row order.

vector

The vector sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Creates the read only vector source from the validated inputs required by MappedVectorSourceFactory.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Returns: The storage-specific vector source, with ownership and disposal obligations defined by the returned type and the Create contract.

Initializes a new LlamaGeneratedToken instance with validated dependencies and operational bounds.

sequence

The zero-based selection sequence within the current generation operation.

tokenId

The token identifier to process; it must fall within the validated vocabulary and operation-specific range.

selectedLogit

The deterministic selected logit at this generation sequence position. NaN is rejected; infinities remain valid because compares them using the same deterministic ordering as finite values.

Reads the local real-model smoke configuration from environment variables.

stage

The explicitly selected real-model smoke stage or stage implementation executed by the controlled pipeline.

Returns: The local smoke options, with ownership and disposal obligations defined by the returned type and the Load contract.

Resolves and validates one local model path.

path

The local file-system path processed by this operation; it must satisfy the containing component's path and scope policy.

allowedRoot

The normalized caller-authorized directory boundary, or null when no containment root was configured for the operation.

maximumFileByteCount

The optional explicit file-size limit; zero disables it.

Returns: The text produced by RealModelPathPolicy.Resolve for this contract: Resolves and validates one local model path. The returned string is detached from mutable caller storage and is not persisted by the operation.

Runs the requested local smoke stages in their required order.

options

The explicit local smoke options and evidence boundaries.

Returns: A bounded machine-readable artifact describing the deepest completed stage.

Creates a SHA-256 fingerprint from the configuration fields that determine cache geometry and semantics.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

Returns: The text produced by ReferenceKvCacheFingerprint.Create for this contract: Creates a SHA-256 fingerprint from the configuration fields that determine cache geometry and semantics. The returned string is detached from mutable caller storage and is not persisted by the operation.

Appends or replaces one layer's key and value vectors at a sequence position.

layerIndex

The zero-based layer index; it must identify an existing position within the relevant validated range.

position

The zero-based sequence or cache position addressed by the operation; it must lie within the allocated context and readable or writable range.

key

The flattened key vector for all key/value heads.

value

The flattened value vector for all key/value heads.

Retrieves the key from the current cache state after validating the requested access.

layerIndex

The zero-based layer index; it must identify an existing position within the relevant validated range.

position

The zero-based sequence or cache position addressed by the operation; it must lie within the allocated context and readable or writable range.

headIndex

The zero-based head index; it must identify an existing position within the relevant validated range.

Returns: The bounded ReadOnlySpan<float> view produced by IReferenceKvCache.GetKey: Retrieves the key from the current cache state after validating the requested access. Its lifetime and ownership remain tied to the owner identified by the containing type; no out-of-range region is exposed.

Retrieves the value from the current cache state after validating the requested access.

layerIndex

The zero-based layer index; it must identify an existing position within the relevant validated range.

position

The zero-based sequence or cache position addressed by the operation; it must lie within the allocated context and readable or writable range.

headIndex

The zero-based head index; it must identify an existing position within the relevant validated range.

Returns: The bounded ReadOnlySpan<float> view produced by IReferenceKvCache.GetValue: Retrieves the value from the current cache state after validating the requested access. Its lifetime and ownership remain tied to the owner identified by the containing type; no out-of-range region is exposed.

Resets the requested state to its validated initial state without publishing partial state.

Creates a bounded snapshot for tiny-fixture testing and replay.

Returns: The immutable cache snapshot, with ownership and disposal obligations defined by the returned type and the CreateSnapshot contract.

Restores the supplied snapshot from a validated persisted representation.

snapshot

The immutable state snapshot being validated, serialized, restored, or analyzed without retaining caller-owned mutable aliases.

Initializes a reference key/value cache with a geometry-derived compatibility fingerprint.

layerCount

The layer count used to bound this operation; it must be nonnegative and within the supported range.

contextLength

The context length that supplies session-scoped identity and boundary state; it is validated before dependent work begins.

keyValueHeadCount

The number of key/value heads per layer.

headWidth

The positive number of scalar values stored for each attention head in one cache position.

Initializes a reference key/value cache with an explicit model/configuration fingerprint.

layerCount

The layer count used to bound this operation; it must be nonnegative and within the supported range.

contextLength

The context length that supplies session-scoped identity and boundary state; it is validated before dependent work begins.

keyValueHeadCount

The number of key/value heads per layer.

headWidth

The positive number of scalar values stored for each attention head in one cache position.

configurationFingerprint

The configuration fingerprint text consumed by ReferenceKvCache.ReferenceKvCache; null, emptiness, length, encoding, identifier, or path rules are enforced as documented, and the value is not persisted by this operation.

Writes the supplied layer index to the current cache state using the component's canonical representation.

layerIndex

The zero-based layer index; it must identify an existing position within the relevant validated range.

position

The zero-based sequence or cache position addressed by the operation; it must lie within the allocated context and readable or writable range.

key

The key sequence used by this operation; its required length, ordering, and element bounds are validated before access.

value

The value sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Retrieves the key from the current cache state after validating the requested access.

layerIndex

The zero-based layer index; it must identify an existing position within the relevant validated range.

position

The zero-based sequence or cache position addressed by the operation; it must lie within the allocated context and readable or writable range.

headIndex

The zero-based head index; it must identify an existing position within the relevant validated range.

Returns: The bounded ReadOnlySpan<float> view produced by ReferenceKvCache.GetKey: Retrieves the key from the current cache state after validating the requested access. Its lifetime and ownership remain tied to the owner identified by the containing type; no out-of-range region is exposed.

Retrieves the value from the current cache state after validating the requested access.

layerIndex

The zero-based layer index; it must identify an existing position within the relevant validated range.

position

The zero-based sequence or cache position addressed by the operation; it must lie within the allocated context and readable or writable range.

headIndex

The zero-based head index; it must identify an existing position within the relevant validated range.

Returns: The bounded ReadOnlySpan<float> view produced by ReferenceKvCache.GetValue: Retrieves the value from the current cache state after validating the requested access. Its lifetime and ownership remain tied to the owner identified by the containing type; no out-of-range region is exposed.

Resets the reference KV cache contents and logical sequence position to their initial state.

Creates the snapshot from the validated inputs required by ReferenceKvCache.

Returns: The ReferenceKvCacheSnapshot result produced by ReferenceKvCache.CreateSnapshot for this contract: Creates the snapshot from the validated inputs required by ReferenceKvCache. It is published only after all documented validation and ownership transitions succeed.

Creates a small diagnostic snapshot without exposing mutable key/value arrays.

Returns: The bounded diagnostic snapshot, with ownership and disposal obligations defined by the returned type and the CreateDiagnosticSnapshot contract.

Restores the supplied snapshot from a validated persisted representation.

snapshot

The immutable state snapshot being validated, serialized, restored, or analyzed without retaining caller-owned mutable aliases.

Serializes a bounded cache snapshot in deterministic little-endian form.

snapshot

The immutable state snapshot being validated, serialized, restored, or analyzed without retaining caller-owned mutable aliases.

modelArtifactFingerprint

The stable model-artifact fingerprint that binds the serialized state to the exact reviewed model identity.

cacheLayoutFingerprint

The stable cache-layout fingerprint used to reject state created for incompatible tensor geometry or storage layout.

maximumByteCount

The maximum byte count used to bound this operation; it must be nonnegative and within the supported range.

Returns: The serialized snapshot bytes including a trailing SHA-256.

Deserializes and verifies a portable key/value-cache snapshot.

bytes

The bytes sequence used by this operation; its required length, ordering, and element bounds are validated before access.

maximumByteCount

The maximum byte count used to bound this operation; it must be nonnegative and within the supported range.

Returns: The ReferenceKvPortableSnapshot result produced by ReferenceKvCacheSerializer.Deserialize for this contract: Deserializes and verifies a portable key/value-cache snapshot. It is published only after all documented validation and ownership transitions succeed.

Restores verified portable bytes into a cache after validating model and layout identities.

cache

The validated ReferenceKvCache dependency consumed by Restore; ownership and lifetime remain with the caller unless this member explicitly documents a transfer.

bytes

The bytes sequence used by this operation; its required length, ordering, and element bounds are validated before access.

expectedModelArtifactFingerprint

The expected artifact identity, or an empty string to accept an empty serialized identity.

expectedCacheLayoutFingerprint

The required cache-layout fingerprint against which the serialized state is compared before restoration.

Copies every vector value into a caller-owned float32 destination.

destination

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Copies and, when required, dequantizes one logical row into a caller-owned float32 destination.

rowIndex

The zero-based row index; it must identify an existing position within the relevant validated range.

destination

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Multiplies this matrix by a float32 vector without materializing a complete float32 matrix.

vector

The input vector with at least ColumnCount values.

output

The output buffer with at least RowCount values.

Initializes a new ArrayVectorSource instance with validated dependencies and operational bounds.

tensorName

The exact ordinal GGUF tensor catalog name used for lookup and diagnostics.

values

The values sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Copies the to into caller-owned storage after validating the requested range and capacity.

destination

The destination buffer that receives the produced values.

Initializes a new ArrayMatrixSource instance with validated dependencies and operational bounds.

tensorName

The exact ordinal GGUF tensor catalog name used for lookup and diagnostics.

values

The values sequence used by this operation; its required length, ordering, and element bounds are validated before access.

rowCount

The row count used to bound this operation; it must be nonnegative and within the supported range.

columnCount

The column count used to bound this operation; it must be nonnegative and within the supported range.

Copies the row to into caller-owned storage after validating the requested range and capacity.

rowIndex

The zero-based row index; it must identify an existing position within the relevant validated range.

destination

The destination buffer that receives the produced values.

Multiplies the supplied vector by the supplied vector without changing logical row order.

vector

The vector sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Initializes a new ArrayLlamaLayerWeightSource instance with validated dependencies and operational bounds.

blockIndex

The zero-based block index; it must identify an existing position within the relevant validated range.

weights

The validated model-weight source or bound weight set consumed read-only by the deterministic reference operation.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

Creates an array-backed source after validating its complete model contract.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

weights

The validated model-weight source or bound weight set consumed read-only by the deterministic reference operation.

Returns: The array-backed source, with ownership and disposal obligations defined by the returned type and the Create contract.

Initializes a new ArrayLlamaModelWeightSource instance with validated dependencies and operational bounds.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

weights

The validated model-weight source or bound weight set consumed read-only by the deterministic reference operation.

Initializes a new MappedFloat32VectorSource instance with validated dependencies and operational bounds.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Copies the to into caller-owned storage after validating the requested range and capacity.

destination

The destination buffer that receives the produced values.

Initializes a new MappedFloat32MatrixSource instance with validated dependencies and operational bounds.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Copies the row to into caller-owned storage after validating the requested range and capacity.

rowIndex

The zero-based row index; it must identify an existing position within the relevant validated range.

destination

The destination buffer that receives the produced values.

Multiplies the supplied vector by the supplied vector without changing logical row order.

vector

The vector sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Initializes a new MappedQ8_0MatrixSource instance with validated dependencies and operational bounds.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Copies the row to into caller-owned storage after validating the requested range and capacity.

rowIndex

The zero-based row index; it must identify an existing position within the relevant validated range.

destination

The destination buffer that receives the produced values.

Multiplies the supplied vector by the supplied vector without changing logical row order.

vector

The vector sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Initializes a new MappedQ4_0MatrixSource instance with validated dependencies and operational bounds.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Copies the row to into caller-owned storage after validating the requested range and capacity.

rowIndex

The zero-based row index; it must identify an existing position within the relevant validated range.

destination

The destination buffer that receives the produced values.

Multiplies the supplied vector by the supplied vector without changing logical row order.

vector

The vector sequence used by this operation; its required length, ordering, and element bounds are validated before access.

output

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

Creates a direct mapped source for supported scalar and quantized storage.

view

The bounded tensor view whose descriptor, shape, byte order, and mapped payload are read without transferring ownership of the underlying mapping.

Returns: The storage-specific matrix source, with ownership and disposal obligations defined by the returned type and the Create contract.

Initializes a new MappedLlamaLayerWeightSource instance with validated dependencies and operational bounds.

weights

The validated model-weight source or bound weight set consumed read-only by the deterministic reference operation.

Creates and validates a complete mapped model weight source.

weights

The validated model-weight source or bound weight set consumed read-only by the deterministic reference operation.

Returns: The validated mapped model weight source, with ownership and disposal obligations defined by the returned type and the Create contract.

Initializes a new MappedLlamaModelWeightSource instance with validated dependencies and operational bounds.

weights

The validated model-weight source or bound weight set consumed read-only by the deterministic reference operation.

Validates every global and block-local source against the configured model geometry.

config

The validated LLaMA model configuration defining context capacity, vocabulary size, tensor geometry, and attention dimensions for the operation.

weights

The validated model-weight source or bound weight set consumed read-only by the deterministic reference operation.

Renders messages using a small role/content template rather than a general Jinja interpreter.

messages

The messages sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Returns: The text produced by ChatTemplateRenderer.Render for this contract: Renders messages using a small role/content template rather than a general Jinja interpreter. The returned string is detached from mutable caller storage and is not persisted by the operation.

Computes a canonical tokenizer fingerprint without treating decoded text as token-ID authority.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

Returns: The text produced by GgufTokenizerFingerprint.Create for this contract: Computes a canonical tokenizer fingerprint without treating decoded text as token-ID authority. The returned string is detached from mutable caller storage and is not persisted by the operation.

Enumerates the known token identifiers in stable source order without exposing mutable internal storage.

Returns: An ordered sequence containing the known token identifiers as produced by the validated operation.

Creates the canonical detokenization options consumed by the tokenizer engine.

Returns: A new canonical options instance with the same behavior flags.

Encodes the supplied text with the configured tokenizer and validated special-token policy.

text

The text processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

options

The optional TokenizationOptions controlling Encode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The TokenizationResult result produced by IGgufTokenizer.Encode for this contract: Encodes the supplied text with the configured tokenizer and validated special-token policy. It is published only after all documented validation and ownership transitions succeed.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The ordered token identifiers to process; sequence order is preserved and each identifier is validated where required.

options

The optional DetokenizationOptions controlling Decode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Reads tokenizer metadata without throwing for validation failures.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

Returns: The GgufTokenizerMetadata result produced by GgufTokenizerMetadataReader.Read for this contract: Reads tokenizer metadata without throwing for validation failures. It is published only after all documented validation and ownership transitions succeed.

Reads tokenizer metadata and throws when validation fails.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

Returns: The GgufTokenizerMetadata result produced by GgufTokenizerMetadataReader.ReadStrict for this contract: Reads tokenizer metadata and throws when validation fails. It is published only after all documented validation and ownership transitions succeed.

Validates the supplied metadata against the invariants required by GgufTokenizerMetadataValidator.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

Returns: The GgufTokenizerMetadataValidationResult result produced by GgufTokenizerMetadataValidator.Validate for this contract: Validates the supplied metadata against the invariants required by GgufTokenizerMetadataValidator. It is published only after all documented validation and ownership transitions succeed.

Initializes a new InvalidGgufTokenizerException instance with validated dependencies and operational bounds.

message

The display-safe diagnostic message describing the failure without embedding prompt text, generated text, credentials, or private file contents.

Initializes a new UnsupportedTokenizerException instance with validated dependencies and operational bounds.

message

The display-safe diagnostic message describing the failure without embedding prompt text, generated text, credentials, or private file contents.

Initializes a GPT-2 BPE engine from validated GGUF tokenizer metadata.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

Encodes the raw with the configured tokenizer and validated special-token policy.

text

The text to process using the configured encoding and normalization rules.

context

The context that supplies session-scoped identity and boundary state; it is validated before dependent work begins.

destination

The destination buffer that receives the produced values.

trace

The trace sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The token identifiers to process in sequence order.

options

The optional DetokenizationOptions controlling Decode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Initializes a tokenizer from validated GGUF metadata.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

Initializes a tokenizer with an explicitly selected family engine.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

engine

The validated IGgufTokenizerEngine dependency consumed by MetadataDrivenGgufTokenizer; ownership and lifetime remain with the caller unless this member explicitly documents a transfer.

Tokenizes the supplied text with the configured metadata and preserves deterministic token order.

text

The text to process using the configured encoding and normalization rules.

Returns: An ordered read-only collection of token text values produced by the configured tokenizer.

Encodes the supplied text with the configured tokenizer and validated special-token policy.

text

The text to process using the configured encoding and normalization rules.

addBos

A value indicating whether add BOS applies to this operation.

addEos

A value indicating whether add EOS applies to this operation.

Returns: An ordered read-only collection of token identifiers produced by the configured tokenizer.

Encodes the supplied text with the configured tokenizer and validated special-token policy.

text

The text to process using the configured encoding and normalization rules.

options

The optional TokenizationOptions controlling Encode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The TokenizationResult result produced by MetadataDrivenGgufTokenizer.Encode for this contract: Encodes the supplied text with the configured tokenizer and validated special-token policy. It is published only after all documented validation and ownership transitions succeed.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The token identifiers to process in sequence order.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The token identifiers to process in sequence order.

options

The optional DetokenizationOptions controlling Decode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Decodes model token identifiers using the stable metadata-driven compatibility options contract.

tokenIds

The token identifiers to process in sequence order.

options

The optional MetadataDrivenGgufTokenizerDetokenizationOptions controlling Decode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Counts the tokens using the same deterministic rules as the corresponding processing operation.

text

The text to process using the configured encoding and normalization rules.

Returns: The int value computed by MetadataDrivenGgufTokenizer.CountTokens for this contract: Counts the tokens using the same deterministic rules as the corresponding processing operation. Range, finite-value, and overflow checks are completed before the value is returned.

Counts the tokens using the same deterministic rules as the corresponding processing operation.

messages

The messages sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Returns: The TokenCountResult result produced by MetadataDrivenGgufTokenizer.CountTokens for this contract: Counts the tokens using the same deterministic rules as the corresponding processing operation. It is published only after all documented validation and ownership transitions succeed.

Initializes the engine from validated GGUF tokenizer metadata.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

Encodes the raw with the configured tokenizer and validated special-token policy.

text

The text to process using the configured encoding and normalization rules.

context

The context that supplies session-scoped identity and boundary state; it is validated before dependent work begins.

destination

The destination buffer that receives the produced values.

trace

The trace sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The token identifiers to process in sequence order.

options

The optional DetokenizationOptions controlling Decode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Creates a raw-text fragment representing an unmodified source slice at the supplied offset.

text

The text processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

offset

The zero-based offset into the relevant source or destination; range validation occurs before access.

Returns: The SpecialTokenFragment result produced by SpecialTokenFragment.Raw for this contract: Creates a raw-text fragment representing an unmodified source slice at the supplied offset. It is published only after all documented validation and ownership transitions succeed.

Creates a special-token fragment at the supplied source-text offset.

tokenId

The token identifier to process; it must fall within the validated vocabulary and operation-specific range.

text

The text processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

offset

The zero-based offset into the relevant source or destination; range validation occurs before access.

Returns: The SpecialTokenFragment result produced by SpecialTokenFragment.Token for this contract: Creates a special-token fragment at the supplied source-text offset. It is published only after all documented validation and ownership transitions succeed.

Partitions text around known special tokens using longest-token-first matching.

text

The text processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

parseSpecial

Whether control and unknown tokens should be parsed as special tokens.

Returns: An ordered read-only IReadOnlyList<SpecialTokenFragment> result from SpecialTokenPartitioner.Partition: Partitions text around known special tokens using longest-token-first matching. Mutable internal collection aliases are not exposed through the returned contract.

Encodes a raw-text fragment into model token identifiers.

text

The text processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

context

The context that supplies session-scoped identity and boundary state; it is validated before dependent work begins.

destination

The caller-owned destination buffer that receives the result; required capacity is validated before any write occurs.

trace

The trace sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The ordered token identifiers to process; sequence order is preserved and each identifier is validated where required.

options

The optional DetokenizationOptions controlling Decode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Attempts to parse the Boolean result while reporting invalid input without a successful result.

text

The text processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

rule

When the method returns, contains the rule produced by the operation when successful; otherwise contains the type's default value.

Returns: True when the rule contains two non-empty symbols.

Determines whether a pre-tokenizer identifier is supported.

name

The exact ordinal name used for catalog lookup, canonical hashing, or diagnostic labeling as defined by the containing member.

Returns: True when the identifier is absent or explicitly supported.

Retrieves the supported names from the configured tokenizer after validating the requested access.

Returns: An ordered read-only IReadOnlyList<string> result from GgufPreTokenizerRegistry.GetSupportedNames: Retrieves the supported names from the configured tokenizer after validating the requested access. Mutable internal collection aliases are not exposed through the returned contract.

Creates the GGUF tokenizer engine from the validated inputs required by GgufTokenizerEngineFactory.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

Returns: The concrete tokenizer engine, with ownership and disposal obligations defined by the returned type and the Create contract.

Creates a tokenizer for a parsed GGUF model.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

Returns: The tokenizer selected from metadata, with ownership and disposal obligations defined by the returned type and the Create contract.

Creates the tokenizer from the validated inputs required by GgufTokenizerFactory.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

Returns: The ITokenizer result produced by GgufTokenizerFactory.Create for this contract: Creates the tokenizer from the validated inputs required by GgufTokenizerFactory. It is published only after all documented validation and ownership transitions succeed.

Creates a tokenizer after strict GGUF tokenizer metadata validation.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

Returns: The selected tokenizer, with ownership and disposal obligations defined by the returned type and the CreateStrict contract.

Initializes the tokenizer from validated GGUF metadata.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

Tokenizes the supplied text with the configured metadata and preserves deterministic token order.

text

The text to process using the configured encoding and normalization rules.

Returns: An ordered read-only collection of token text values produced by the configured tokenizer.

Encodes the supplied text with the configured tokenizer and validated special-token policy.

text

The text to process using the configured encoding and normalization rules.

addBos

A value indicating whether add BOS applies to this operation.

addEos

A value indicating whether add EOS applies to this operation.

Returns: An ordered read-only collection of token identifiers produced by the configured tokenizer.

Encodes the supplied text with the configured tokenizer and validated special-token policy.

text

The text to process using the configured encoding and normalization rules.

options

The optional TokenizationOptions controlling Encode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The TokenizationResult result produced by SentencePieceGgufTokenizer.Encode for this contract: Encodes the supplied text with the configured tokenizer and validated special-token policy. It is published only after all documented validation and ownership transitions succeed.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The token identifiers to process in sequence order.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The token identifiers to process in sequence order.

options

The optional DetokenizationOptions controlling Decode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Counts the tokens using the same deterministic rules as the corresponding processing operation.

text

The text to process using the configured encoding and normalization rules.

Returns: The int value computed by SentencePieceGgufTokenizer.CountTokens for this contract: Counts the tokens using the same deterministic rules as the corresponding processing operation. Range, finite-value, and overflow checks are completed before the value is returned.

Counts the tokens using the same deterministic rules as the corresponding processing operation.

messages

The messages sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Returns: The TokenCountResult result produced by SentencePieceGgufTokenizer.CountTokens for this contract: Counts the tokens using the same deterministic rules as the corresponding processing operation. It is published only after all documented validation and ownership transitions succeed.

Initializes the tokenizer from validated GGUF metadata.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

Tokenizes the supplied text with the configured metadata and preserves deterministic token order.

text

The text to process using the configured encoding and normalization rules.

Returns: An ordered read-only collection of token text values produced by the configured tokenizer.

Encodes the supplied text with the configured tokenizer and validated special-token policy.

text

The text to process using the configured encoding and normalization rules.

addBos

A value indicating whether add BOS applies to this operation.

addEos

A value indicating whether add EOS applies to this operation.

Returns: An ordered read-only collection of token identifiers produced by the configured tokenizer.

Encodes the supplied text with the configured tokenizer and validated special-token policy.

text

The text to process using the configured encoding and normalization rules.

options

The optional TokenizationOptions controlling Encode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The TokenizationResult result produced by Gpt2BpeTokenizer.Encode for this contract: Encodes the supplied text with the configured tokenizer and validated special-token policy. It is published only after all documented validation and ownership transitions succeed.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The token identifiers to process in sequence order.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The token identifiers to process in sequence order.

options

The optional DetokenizationOptions controlling Decode; null selects the documented defaults, supplied limits are validated before allocation, and the instance is not mutated.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Counts the tokens using the same deterministic rules as the corresponding processing operation.

text

The text to process using the configured encoding and normalization rules.

Returns: The int value computed by Gpt2BpeTokenizer.CountTokens for this contract: Counts the tokens using the same deterministic rules as the corresponding processing operation. Range, finite-value, and overflow checks are completed before the value is returned.

Counts the tokens using the same deterministic rules as the corresponding processing operation.

messages

The messages sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Returns: The TokenCountResult result produced by Gpt2BpeTokenizer.CountTokens for this contract: Counts the tokens using the same deterministic rules as the corresponding processing operation. It is published only after all documented validation and ownership transitions succeed.

Tokenizes the supplied text with the configured metadata and preserves deterministic token order.

text

The text to process using the configured encoding and normalization rules.

Returns: An ordered read-only collection of token text values produced by the configured tokenizer.

Encodes the supplied text with the configured tokenizer and validated special-token policy.

text

The text to process using the configured encoding and normalization rules.

addBos

A value indicating whether add BOS applies to this operation.

addEos

A value indicating whether add EOS applies to this operation.

Returns: An ordered read-only collection of token identifiers produced by the configured tokenizer.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The token identifiers to process in sequence order.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Counts the tokens using the same deterministic rules as the corresponding processing operation.

text

The text to process using the configured encoding and normalization rules.

Returns: The int value computed by RwkvWorldTokenizer.CountTokens for this contract: Counts the tokens using the same deterministic rules as the corresponding processing operation. Range, finite-value, and overflow checks are completed before the value is returned.

Counts the tokens using the same deterministic rules as the corresponding processing operation.

messages

The messages sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Returns: The TokenCountResult result produced by RwkvWorldTokenizer.CountTokens for this contract: Counts the tokens using the same deterministic rules as the corresponding processing operation. It is published only after all documented validation and ownership transitions succeed.

Creates a tokenizer from embedded tokenizer JSON metadata when supported.

json

The json text consumed by HuggingFaceTokenizerJsonAdapter.Create; null, emptiness, length, encoding, identifier, or path rules are enforced as documented, and the value is not persisted by this operation.

Returns: A tokenizer instance, with ownership and disposal obligations defined by the returned type and the Create contract.

Renders and validates a chat template against a message sequence.

template

The template text. The current safe subset ignores arbitrary code.

messages

The messages sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Returns: The TokenizerParityReport result produced by ChatTemplateConformanceSuite.RenderAndValidate for this contract: Renders and validates a chat template against a message sequence. It is published only after all documented validation and ownership transitions succeed.

Truncates messages so the total token count does not exceed the budget.

messages

The messages sequence used by this operation; its required length, ordering, and element bounds are validated before access.

tokenizer

The validated ITokenizer dependency consumed by TruncateMessages; ownership and lifetime remain with the caller unless this member explicitly documents a transfer.

maxTokens

The numeric max tokens consumed by TruncateMessages; it must satisfy the member's documented range, geometry, and finite-value requirements.

Returns: An ordered read-only IReadOnlyList<LlmMessage> result from TokenBudgetTruncator.TruncateMessages: Truncates messages so the total token count does not exceed the budget. Mutable internal collection aliases are not exposed through the returned contract.

Loads ordered tokenizer golden record collection from a verified local source into TokenizerGoldenCorpus.

json

The json text consumed by TokenizerGoldenCorpus.Load; null, emptiness, length, encoding, identifier, or path rules are enforced as documented, and the value is not persisted by this operation.

Returns: An ordered read-only IReadOnlyList<TokenizerGoldenRecord> result from TokenizerGoldenCorpus.Load: Loads ordered tokenizer golden record collection from a verified local source into TokenizerGoldenCorpus. Mutable internal collection aliases are not exposed through the returned contract.

Validates a managed string and optionally replaces unpaired surrogate code units.

text

The text processed by the configured encoding or normalization rules; it must satisfy the declared nullability contract.

policy

The policy that define validation limits and execution behavior; required values are checked before use.

Returns: The original string when valid, or a normalized replacement string when requested.

Decodes one byte fragment and retains incomplete UTF-8 state for the next call.

bytes

The bytes sequence used by this operation; its required length, ordering, and element bounds are validated before access.

flush

The flush flag selecting the documented branch of Decode; it does not grant authority beyond this in-memory operation.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Resets the requested state to its validated initial state without publishing partial state.

Validates embedded tokenizer JSON without allowing it to reorder GGUF token identifiers.

metadata

The metadata containing validated format or tokenizer metadata required by this operation.

Returns: The TokenizerVocabularyReconciliationResult result produced by TokenizerVocabularyReconciler.Reconcile for this contract: Validates embedded tokenizer JSON without allowing it to reorder GGUF token identifiers. It is published only after all documented validation and ownership transitions succeed.

Tokenizes the supplied text with the configured metadata and preserves deterministic token order.

text

The text to process using the configured encoding and normalization rules.

Returns: An ordered read-only collection of token text values produced by the configured tokenizer.

Encodes the supplied text with the configured tokenizer and validated special-token policy.

text

The text to process using the configured encoding and normalization rules.

addBos

A value indicating whether add BOS applies to this operation.

addEos

A value indicating whether add EOS applies to this operation.

Returns: An ordered read-only collection of token identifiers produced by the configured tokenizer.

Decodes the supplied token sequence with the configured tokenizer while preserving sequence order.

tokenIds

The token identifiers to process in sequence order.

Returns: The decoded text produced from the validated token sequence in the original sequence order.

Counts the tokens using the same deterministic rules as the corresponding processing operation.

text

The text to process using the configured encoding and normalization rules.

Returns: The int value computed by WhitespaceTokenizer.CountTokens for this contract: Counts the tokens using the same deterministic rules as the corresponding processing operation. Range, finite-value, and overflow checks are completed before the value is returned.

Counts the tokens using the same deterministic rules as the corresponding processing operation.

messages

The messages sequence used by this operation; its required length, ordering, and element bounds are validated before access.

Returns: The TokenCountResult result produced by WhitespaceTokenizer.CountTokens for this contract: Counts the tokens using the same deterministic rules as the corresponding processing operation. It is published only after all documented validation and ownership transitions succeed.

Initializes a new GgufFormatException instance with validated dependencies and operational bounds.

code

The stable machine-readable diagnostic code used to classify the failure without relying on localized message text.

message

The display-safe diagnostic message describing the failure without embedding prompt text, generated text, credentials, or private file contents.

byteOffset

The zero-based byte offset into the relevant source or destination; range validation occurs before access.

innerException

The underlying exception preserved for diagnostic chaining, or null when no lower-level failure is available.

Creates a textual dump for a parsed GGUF model.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

Returns: The text produced by GgufDumpWriter.Write for this contract: Creates a textual dump for a parsed GGUF model. The returned string is detached from mutable caller storage and is not persisted by the operation.

Tries to read a Boolean metadata value.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

key

The exact ordinal metadata or catalog key used for lookup; the operation does not normalize or retain a mutable alias of the key.

value

When the method returns, contains the value produced by the operation when successful; otherwise contains the type's default value.

Returns: True when the key exists and contains a Boolean value.

Tries to read a metadata string array.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

key

The exact ordinal metadata or catalog key used for lookup; the operation does not normalize or retain a mutable alias of the key.

values

When the method returns, contains the values produced by the operation when successful; otherwise contains the type's default value.

Returns: True when the key exists and contains only string elements.

Tries to read a metadata single-precision floating point array.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

key

The exact ordinal metadata or catalog key used for lookup; the operation does not normalize or retain a mutable alias of the key.

values

When the method returns, contains the values produced by the operation when successful; otherwise contains the type's default value.

Returns: True when the key exists and contains numeric elements convertible to float.

Tries to read a metadata 32-bit signed integer array.

model

The parsed GGUF model whose validated metadata and tensor catalog are consumed by this operation.

key

The exact ordinal metadata or catalog key used for lookup; the operation does not normalize or retain a mutable alias of the key.

values

When the method returns, contains the values produced by the operation when successful; otherwise contains the type's default value.

Returns: True when the key exists and contains integral elements convertible to int.

Tries to read a metadata byte array.