decision-toolkit 0.4.3__tar.gz → 0.4.4__tar.gz

{decision_toolkit-0.4.3 → decision_toolkit-0.4.4}/CHANGELOG.md

@@ -1,8 +1,15 @@
 # Changelog
 
-All notable changes to this project are documented here. Version `0.4.3` is
+All notable changes to this project are documented here. Version `0.4.4` is
 published on PyPI.
 
+## [0.4.4] — 2026-10-21
+
+Documentation correction: removed the remaining wording that described
+`variant` as a request extension. System One and OpenAI request bodies have no
+`variant` field; quantization is selected with the model alias and the chosen
+variant is reported under `x_decision_toolkit` in responses. No runtime changes.
+
 ## [0.4.3] — 2026-10-21
 
 Documentation and help UX improvements. Added a Quickstart, Installation guide,
@@ -235,6 +242,7 @@ README correction. No code or model behavior changed.
 - Typed `DecisionModel` API with lifecycle, batching, async, usage, warnings,
   normalized confidence, capabilities, structured events and atomic model store.
 
+[0.4.4]: https://pypi.org/project/decision-toolkit/0.4.4/
 [0.4.3]: https://pypi.org/project/decision-toolkit/0.4.3/
 [0.4.2]: https://pypi.org/project/decision-toolkit/0.4.2/
 [0.4.1]: https://pypi.org/project/decision-toolkit/0.4.1/

{decision_toolkit-0.4.3 → decision_toolkit-0.4.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: decision-toolkit
-Version: 0.4.3
+Version: 0.4.4
 Summary: Local-first toolkit for open-weights decision models: one ONNX engine, one adapter per model contract, typed choice/yes-no/score decisions with probabilities on CPU.
 Project-URL: Homepage, https://github.com/gia-uh/decision-toolkit
 Project-URL: Repository, https://github.com/gia-uh/decision-toolkit
@@ -70,7 +70,7 @@ uv pip install "decision-toolkit[onnx,server]" # + REST server (FastAPI)
 The project is managed with [uv](https://docs.astral.sh/uv/) and ships a committed `uv.lock`. Published
 wheels remain compatible with `pip` for environments that do not use uv.
 
-PyPI: <https://pypi.org/project/decision-toolkit/0.4.3/>. See
+PyPI: <https://pypi.org/project/decision-toolkit/0.4.4/>. See
 [docs/RELEASE.md](https://github.com/gia-uh/decision-toolkit/blob/main/docs/RELEASE.md)
 for the Trusted Publishing flow and clean-install checks.
 

{decision_toolkit-0.4.3 → decision_toolkit-0.4.4}/README.md

@@ -32,7 +32,7 @@ uv pip install "decision-toolkit[onnx,server]" # + REST server (FastAPI)
 The project is managed with [uv](https://docs.astral.sh/uv/) and ships a committed `uv.lock`. Published
 wheels remain compatible with `pip` for environments that do not use uv.
 
-PyPI: <https://pypi.org/project/decision-toolkit/0.4.3/>. See
+PyPI: <https://pypi.org/project/decision-toolkit/0.4.4/>. See
 [docs/RELEASE.md](https://github.com/gia-uh/decision-toolkit/blob/main/docs/RELEASE.md)
 for the Trusted Publishing flow and clean-install checks.
 

{decision_toolkit-0.4.3 → decision_toolkit-0.4.4}/docs/RELEASE.md

@@ -13,7 +13,8 @@ changed the public documentation URLs to this repository's Pages site; `0.3.0`
 added the Laya multilingual, Open-Jev DeBERTa v3 large and Strands Decider ONNX models; `0.3.1` made the catalog list one row per variant with exact commands; `0.4.0`
 replaced the non-standard REST `variant` body field with standard model aliases
 (`id@variant`); `0.4.3` added the user-facing documentation guides and cleaned
-up CLI help/status output.
+up CLI help/status output; `0.4.4` corrected the remaining REST wording about
+variant request fields.
 
 ## Local build and verification
 
@@ -40,7 +41,7 @@ Clean install test (no `PYTHONPATH`):
 ```bash
 uv venv --python 3.12 /tmp/dtk-wheel
 uv pip install --python /tmp/dtk-wheel/bin/python \
-  dist/decision_toolkit-0.4.3-py3-none-any.whl
+  dist/decision_toolkit-0.4.4-py3-none-any.whl
 uv pip check --python /tmp/dtk-wheel/bin/python
 /tmp/dtk-wheel/bin/dtk --version
 /tmp/dtk-wheel/bin/dtk models list --json
@@ -51,7 +52,7 @@ With inference and server extras:
 
 ```bash
 uv pip install --python /tmp/dtk-wheel/bin/python \
-  "dist/decision_toolkit-0.4.3-py3-none-any.whl[onnx,server,tracing]"
+  "dist/decision_toolkit-0.4.4-py3-none-any.whl[onnx,server,tracing]"
 /tmp/dtk-wheel/bin/dtk decide --model gliner2.5-decide ...
 /tmp/dtk-wheel/bin/dtk serve --model gliner2.5-decide --warmup --port 8016
 curl -s http://127.0.0.1:8016/health
@@ -85,8 +86,8 @@ No long-lived PyPI token is needed; the workflow uses GitHub OIDC
 4. Create and push an annotated tag:
 
    ```bash
-   git tag -a v0.4.3 -m "decision-toolkit 0.4.3"
-   git push origin v0.4.3
+   git tag -a v0.4.4 -m "decision-toolkit 0.4.4"
+   git push origin v0.4.4
    ```
 
 5. `.github/workflows/release.yml` builds once, publishes to **TestPyPI**, then
@@ -135,9 +136,9 @@ to TestPyPI without publishing to PyPI or creating a GitHub release.
 
 ## Published release
 
-`0.4.3` is published through Trusted Publishing on 2026-10-21:
+`0.4.4` is published through Trusted Publishing on 2026-10-21:
 
-<https://pypi.org/project/decision-toolkit/0.4.3/>
+<https://pypi.org/project/decision-toolkit/0.4.4/>
 
 `0.1.0` was the first release; `0.1.1` refreshed the PyPI description; `0.1.2`
 validated the OIDC TestPyPI → PyPI flow; `0.1.3` added post-publish install
@@ -151,7 +152,8 @@ added the three new ONNX decision models; `0.3.1` made the catalog list one row
 selection to standard model aliases; `0.4.1` exposed aliases in the catalog and
 per-variant loading examples in every model card; `0.4.2` removed internal
 project language and public proposal/roadmap pages; `0.4.3` added user-facing
-guides and cleaned up CLI help/status output.
+guides and cleaned up CLI help/status output; `0.4.4` corrected the remaining
+REST variant-request wording.
 
 The tag-triggered workflow remains the preferred path for future releases once
 the TestPyPI/PyPI trusted publishers and protected environments are configured.

{decision_toolkit-0.4.3 → decision_toolkit-0.4.4}/docs/REST.md

@@ -147,7 +147,8 @@ Standard fields:
   current catalog is text-only. Legacy `files`/`images` fields are local
   extensions; valid data URLs also return `501`, malformed values return `422`.
 - `model` is optional only as a local convenience; standard clients send it.
-  `variant` is a local extension, not part of the TypeSafe schema.
+  There is no `variant` request field; the chosen variant is reported as
+  `x_decision_toolkit.variant` in the response.
 
 ## OpenAI Decisions beta
 
@@ -211,7 +212,8 @@ Response:
 - Duplicate question names return `400`.
 - Images in `input` return `501` (`multimodal_not_supported`); the published
   beta supports inline images, but the current catalog has no vision model.
-- `variant` is a local extension; standard clients omit it.
+- `x_decision_toolkit.variant` is a response extension; standard clients ignore
+  it, and the request has no `variant` field.
 - OpenAI SDK clients can list models (`client.models.list()`) and parse the
   error envelope (`{"error": {message, type, param, code}}`).
 
@@ -309,7 +311,8 @@ Adding a new spec version means adding a new module, not editing the old one.
 | Images/assets | `501` for current models | `501` for current models |
 | Usage | `input_tokens` + non-negative `output_tokens` | input/output/total + details |
 | Per-request model | standard field; optional local default | standard field; optional local default |
-| `variant` | local extension | local extension |
+| `variant` request field | not present | not present |
+| `x_decision_toolkit.variant` response | extension | extension |
 | Validation errors | `422` + `detail` | `400` + envelope |
 | Auth / CORS / limits | server-level | server-level |
 

{decision_toolkit-0.4.3 → decision_toolkit-0.4.4}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "decision-toolkit"
-version = "0.4.3"
+version = "0.4.4"
 description = "Local-first toolkit for open-weights decision models: one ONNX engine, one adapter per model contract, typed choice/yes-no/score decisions with probabilities on CPU."
 readme = "README.md"
 requires-python = ">=3.12"

{decision_toolkit-0.4.3 → decision_toolkit-0.4.4}/src/decision_toolkit/__init__.py

@@ -7,7 +7,7 @@ questions with a calibrated probability for every option, on CPU.
 
 from __future__ import annotations
 
-__version__ = "0.4.3"
+__version__ = "0.4.4"
 
 from .adapters import register_adapter
 from .calibration import brier_score, expected_calibration_error, threshold_sweep

{decision_toolkit-0.4.3 → decision_toolkit-0.4.4}/src/decision_toolkit/contracts/systemone_v1.py

@@ -13,7 +13,7 @@ documentation at ``https://docs.typesafe.ai/api``:
 * unsupported image/asset input returns ``501`` because the current catalog is
   text-only
 
-``model`` and ``variant`` remain optional routing extensions for the local
+``model`` remains an optional routing extension for the local
 multi-model server; standard clients send ``model``. Toolkit-specific metadata
 lives under ``x_decision_toolkit`` so standard fields keep their meaning.
 """
@@ -103,9 +103,9 @@ class SystemOneQuestion(BaseModel):
 class SystemOneRequest(BaseModel):
     """System One request body.
 
-    ``model`` and ``variant`` are routing extensions: System One itself does not
-    require a model field, but a multi-model local server does. They are
-    optional and standard clients can omit them.
+    ``model`` is a routing extension: System One itself does not require a model
+    field, but a multi-model local server does. It is optional and standard
+    clients can omit it.
     """
 
     model_config = ConfigDict(extra="forbid")

{decision_toolkit-0.4.3 → decision_toolkit-0.4.4}/uv.lock

@@ -349,7 +349,7 @@ wheels = [
 
 [[package]]
 name = "decision-toolkit"
-version = "0.4.3"
+version = "0.4.4"
 source = { editable = "." }
 dependencies = [
     { name = "huggingface-hub" },