Documentation
¶
Overview ¶
Package llm holds LLM model-identifier helpers shared by the proxy and the management server. The proxy normalizes model ids parsed off inbound requests; management normalizes the operator's registered model ids at synthesis time so both sides of the pricing / routing contract compare equal.
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func NormalizeAnthropicModel ¶ added in v0.78.0
NormalizeAnthropicModel strips the trailing release-date suffix from a Claude model id, e.g. "claude-sonnet-4-5-20250929" -> "claude-sonnet-4-5", so a dated id a client pins matches the undated one the operator registered. Ids that are not Claude-family are returned untouched. Callers try the verbatim id first and fall back to this, so two dated releases of the same family stay distinct wherever both are registered explicitly.
func NormalizeBedrockModel ¶
NormalizeBedrockModel strips an ARN wrapper, a cross-region inference-profile prefix, and the version/throughput suffix from a Bedrock model id so it matches the catalog/pricing key, e.g. "eu.anthropic.claude-sonnet-4-5-20250929-v1:0" -> "anthropic.claude-sonnet-4-5" and the inference-profile ARN's last segment likewise. It is the single source of truth shared by the proxy's request parser (which normalizes the request model from the URL path), the proxy's router (which normalizes the operator's registered Bedrock model ids so both sides compare equal), and the management synthesizer (which keys per-provider pricing entries by the normalized id the parser will emit at billing time).
func NormalizeVertexModel ¶
NormalizeVertexModel strips the "@version" suffix from a Vertex AI model id (e.g. "claude-sonnet-4-5@20250929" -> "claude-sonnet-4-5") so it matches the catalog/pricing key. Vertex publisher models are priced under their vendor surface with the bare, unversioned id.
Types ¶
This section is empty.
Source Files