invera-sdk changelog
All notable changes to the invera-sdk Python package. The Python and
TypeScript SDKs share one version and are released together.
[0.7.0] - 2026-11-05
- Breaking: requires Engine Contract
>=1.3.1 <2.0.0(was>=1.3.0 <2.0.0). Contract 1.3.1 reportssdk_hints.recommended_tool_parserfor each model; a Contract 1.3.0 Engine is rejected withEngineCompatibilityError. - The client now serves any model its invera Engine reports in
GET /v1/models, not only release-pinned profiles. It loads the tokenizer from the reported Hugging Facetokenizer.idand immutabletokenizer.revisionand accepts it only when its locally computed fingerprint equalstokenizer.fingerprint. A new model or tokenizer revision needs no SDK release. model_typeis no longer an allowlist. Discovery is instead checked for a pinned tokenizer identity (40-character commit revision, SHA-256 fingerprint), an available chat template, valid EOS and special token ids, and a supportedsdk_hints.recommended_tool_parser. Each failure raisesEngineCompatibilityErrorbefore anyPOST /v1/raw_attribution; a tokenizer fingerprint mismatch now raisesEngineCompatibilityErrorinstead ofRuntimeError.- Models whose Engine recommends no tool parser stream text completions;
a request with
toolsfor such a model raisesEngineCompatibilityError. - Tool calls are now parsed in both formats the Engine recommends:
qwen_json_tool_call, Hermes-style JSON inside<tool_call>(Qwen2.5, Qwen3, QwQ), andqwen_xml_tool_call, XML per parameter (Qwen3.5 and later). The client picks the format fromsdk_hints.recommended_tool_parser. Streaming chunks,ParameterEvidence, and the pre-execution tool-call decision are equivalent for the same call in either format; a malformed JSON call is reported with the schema issue codeinvalid_json. - Breaking (advanced):
parse_qwen_tool_callsandstream_evidencerequire the tool parser (stream_evidence(..., tool_parser=...)); there is no implicitqwen_xml_tool_calldefault. Pass the Engine'srecommended_tool_parser;stream_evidenceacceptsNonefor streams without tool calls. - Discovery must report exactly one model per Engine endpoint; zero or
several entries raise
EngineCompatibilityError(several were previously resolved to the first supported entry). The SDK still never asks the Engine for a specific model. - A custom
TokenizerAdapterpassed tocreate_client_with_tokenizernow works for any model whose identity exactly matches discovery, without static profile resolution. - Breaking (
invera_sdk.advanced):MODEL_PROFILES,ModelProfile, andSUPPORTED_MODEL_PROFILESare removed; the newSUPPORTED_TOOL_PARSERSlists the parsable tool-parser hints.create_hugging_face_tokenizernow takes the expectedTokenizerIdentityinstead of aModelProfile.
[0.6.1] - 2026-10-02
- Supports invera Engines serving
Qwen/Qwen3.8-27Bthrough the new release-pinnedqwen3.8-27btokenizer profile. The client still picks the profile from the tokenizer identity the Engine reports; there is no model option.
[0.6.0] - 2026-10-02
- Breaking: the SDK now supports only invera Engines serving
Qwen/Qwen3.5-4B(tokenizer profileqwen3.5-4b, model familyqwen3_5).Qwen/Qwen3-4B-Instruct-2507is no longer supported. - Breaking: the
modelclient option is removed. The client always uses the release-pinned model profile; it cannot choose a model. - Breaking: requires Engine Contract
>=1.3.0 <2.0.0(was>=1.2.0 <2.0.0). - Breaking: tool calls are parsed only in the Qwen3.5 XML format
(
<function=…><parameter=…>…</parameter></function>); parameter values are typed from the tool's JSON schema. A malformed call is reported with the new schema issue codeinvalid_xml, which replacesinvalid_json. - The default tokenizer adapter now works with Transformers 4.x as well: its tokenizer fingerprint no longer depends on the installed Transformers major version, which previously failed the pinned-profile check on 4.x.
tool-descriptionmessages are display-only: the default tokenizer never renders them into the prompt. Pass tool definitions throughtools.
[0.5.2] - 2026-09-25
- New
api_keyclient option: a static Engine API key sent asAuthorization: Bearer <key>on every Engine request.
[0.5.0] - 2026-09-22
- First release published to PyPI.
- Licensed under the PolyForm Perimeter License 1.0.1 together with the Layerwise SDK Additional Permission; both texts ship in the distribution.
- The package README is reduced to name, version, vendor, documentation link, license pointer, and this changelog.
[0.4.3] - 2026-09-21
- No Python changes; version kept in lockstep with the TypeScript SDK.
[0.4.2] - 2026-09-21
invera_sdk.__version__added as an alias ofSDK_VERSION.
[0.4.1] - 2026-09-17
- Output words are split at internal whitespace in attribution word mapping.
[0.4.0] - 2026-09-02
- Emitted attribution word mapping aligned between Python and TypeScript.
- Releases are gated by live Engine end-to-end checks.
[0.3.2] - 2026-08-25
- Supports invera Engine Contract 2.x on the v1 routes.
[0.3.1] - 2026-07-30
- Renamed to invera. Golden Path client, evidence stream, and the
invera_sdk.advancedextension module.