From 53f8182e7a4ab41e65b629f7b2df6fe35ff58c8d Mon Sep 17 00:00:00 2001 From: Avi Back To Black <54722547+AviBackToBlack@users.noreply.github.com> Date: Mon, 14 Sep 2026 13:07:40 +0100 Subject: [PATCH 1/4] chore: bump package version to 0.3.0 --- pyproject.toml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/pyproject.toml b/pyproject.toml index 812e947..50d679b 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta" [project] name = "litellm-codex-models" -version = "0.2.1" +version = "0.3.0" description = "Generate a Codex model catalog from LiteLLM /v1/model/info" readme = "README.md" requires-python = ">=3.11" From fbf12ff3c9c2b9f2742186bbafe17e6e4a200533 Mon Sep 17 00:00:00 2001 From: Avi Back To Black <54722547+AviBackToBlack@users.noreply.github.com> Date: Mon, 14 Sep 2026 13:07:47 +0100 Subject: [PATCH 2/4] chore: bump runtime version to 0.3.0 --- src/litellm_codex_models/__init__.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/litellm_codex_models/__init__.py b/src/litellm_codex_models/__init__.py index 5710926..e5b76b7 100644 --- a/src/litellm_codex_models/__init__.py +++ b/src/litellm_codex_models/__init__.py @@ -1,3 +1,3 @@ """LiteLLM -> Codex model catalog generator.""" -__version__ = "0.2.1" +__version__ = "0.3.0" From 815130ea88093aada6e955fbc5689bac0a307430 Mon Sep 17 00:00:00 2001 From: Avi Back To Black <54722547+AviBackToBlack@users.noreply.github.com> Date: Mon, 14 Sep 2026 13:08:01 +0100 Subject: [PATCH 3/4] test: expect CLI version 0.3.0 --- tests/test_cli.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/test_cli.py b/tests/test_cli.py index 8bcc85c..14bffb4 100644 --- a/tests/test_cli.py +++ b/tests/test_cli.py @@ -25,4 +25,4 @@ def test_cli_version_does_not_require_config(capsys): with pytest.raises(SystemExit) as exc_info: build_parser().parse_args(["--version"]) assert exc_info.value.code == 0 - assert capsys.readouterr().out == "litellm-codex-models 0.2.1\n" + assert capsys.readouterr().out == "litellm-codex-models 0.3.0\n" From 84d5255858a051fa1478a7e738edb72a4325bece Mon Sep 17 00:00:00 2001 From: Avi Back To Black <54722547+AviBackToBlack@users.noreply.github.com> Date: Mon, 14 Sep 2026 13:08:42 +0100 Subject: [PATCH 4/4] docs: prepare README for v0.3.0 --- README.md | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 6fe80a7..af1e1fa 100644 --- a/README.md +++ b/README.md @@ -19,10 +19,10 @@ Requires Python 3.11+ and no third-party runtime dependencies. ### Stable release from GitHub -After the `v0.2.0` tag exists, install that exact release directly from GitHub: +After the `v0.3.0` tag exists, install that exact release directly from GitHub: ```bash -python -m pip install "git+https://github.com/AviBackToBlack/litellm-codex-models.git@v0.2.0" +python -m pip install "git+https://github.com/AviBackToBlack/litellm-codex-models.git@v0.3.0" ``` This is the recommended VCS install because it is reproducible. @@ -183,6 +183,13 @@ For an **exact Codex template match**, `context_window` and `max_context_window` For a **foreign model group**, the generator uses the minimum known LiteLLM `max_input_tokens` across every deployment as the best safe approximation for both context fields. A multi-deployment foreign group with missing or invalid context evidence fails closed rather than advertising a guessed window. An explicit `max_input_tokens` override can supply the trusted effective evidence needed to synthesize that group. A single foreign deployment preserves the v0.2 fallback behavior. +## v0.3 highlights + +- Multiple LiteLLM deployments sharing one user-facing `model_name` are aggregated deterministically with conservative safe-to-route capability evidence instead of rejecting duplicates or choosing an arbitrary donor. +- Verified offline Codex bundles bind catalog/prompt/schema resources to one declared repository/ref identity and verify SHA-256 digests before generation. +- Deterministic case-sensitive model globs and exact-keyed trusted `model_overrides` add controlled selection and compatibility correction with full `explain` provenance. +- LiteLLM web-search evidence is explicitly decoupled from Codex `supports_search_tool`; hosted web-search compatibility remains provider/runtime controlled and is not inferred from similarly named metadata. + ## v0.2 highlights - Model-specific Codex donor fields are never inherited by foreign models.