test(verifiers): add Qwen3.8-27B to renderer test matrices - #2416
Conversation
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 2c93838. Configure here.
| _TRUNCATED_ANCHOR_MODELS = [ | ||
| pytest.param("Qwen/Qwen3-8B", "auto", id="Qwen/Qwen3-8B"), | ||
| pytest.param("Qwen/Qwen3.5-9B", "auto", id="Qwen/Qwen3.5-9B"), | ||
| pytest.param("Qwen/Qwen3.8-27B", "auto", id="Qwen/Qwen3.8-27B"), |
There was a problem hiding this comment.
Missing qwen3.8 renderer dependency
Medium Severity
Qwen/Qwen3.8-27B is added with "auto" to exercise a new qwen3.8 renderer, but this repo still locks renderers at 0.1.9.dev9, which only documents qwen3 / qwen3.5 / qwen3.6 / qwen3-vl. Without a matching dependency bump, "auto" cannot resolve to that new renderer, so these matrix entries miss the intended coverage (or land on the wrong hand-coded path / DefaultRenderer).
Additional Locations (1)
Reviewed by Cursor Bugbot for commit 2c93838. Configure here.
ApprovabilityVerdict: Approved at Macroscope's review found this PR approvable — Test-only changes adding a new model to renderer test matrices. No production code modified. The unresolved comment about dependency support affects test coverage effectiveness rather than runtime behavior. Notes:
You can add or adjust custom eligibility rules. Learn more. |


What
Adds Qwen3.8-27B to the renderer end-to-end and client test matrices so the new dense model is covered by the same renderer → RendererClient → Environment wire-protocol tests as the rest of the Qwen family.
Changes
tests/test_renderer_e2e.py— add("Qwen/Qwen3.8-27B", "auto")to_MODEL_FAMILIES.tests/test_renderer_client.py— addQwen/Qwen3.8-27Bto_TRUNCATED_ANCHOR_MODELS.Qwen3.8-27B auto-resolves to the new
qwen3.8renderer (see PrimeIntellect-ai/renderers#132).Testing
test_renderer_e2e.py+test_renderer_client.pyQwen3.8 cases pass.Note
Low Risk
Test-only matrix expansion; no production code or protocol changes.
Overview
Adds
Qwen/Qwen3.8-27B(auto-resolved to theqwen3.8renderer) to the renderer e2e and truncated-anchor client test matrices so the new dense Qwen family is covered by the same TITO wire-protocol and prefix-preserving bridge checks as other Qwen models.Reviewed by Cursor Bugbot for commit 2c93838. Bugbot is set up for automated code reviews on this repo. Configure here.
Note
Add
Qwen/Qwen3.8-27Bto renderer test matricesAdds the
Qwen/Qwen3.8-27Bmodel withautorenderer to_TRUNCATED_ANCHOR_MODELSin test_renderer_client.py and_MODEL_FAMILIESin test_renderer_e2e.py. This expands parameterized test coverage to include the new model.Macroscope summarized 2c93838.