fix(deepseek): wire thinking-mode via DeepSeekProfile (closes #15700, #17212, #17825)#26648
Merged
Conversation
…Seek API DeepSeek's thinking mode requires both: - extra_body.thinking.type: "enabled" to activate thinking mode - top-level reasoning_effort: "max" or "high" to control depth Previously, the ChatCompletionsTransport only handled Kimi's thinking mode — DeepSeek was left unmapped, so reasoning_effort config was silently dropped. This patch: 1. Adds is_deepseek: bool to the Params dataclass, detected by base_url matching api.deepseek.com 2. Maps Hermes effort levels (xhigh/max → "max", low/medium/high → themselves) to the top-level reasoning_effort parameter 3. Sets extra_body.thinking.type alongside the effort 4. Strips reasoning_content from assistant messages sent back to DeepSeek, preventing 400 errors when thinking was enabled
…lback The cherry-picked PR #15251 from @tw2818 correctly identified the DeepSeek 400 root cause but placed the fix in the legacy fallback path of `build_kwargs`, which DeepSeek never reaches — DeepSeek has a registered ProviderProfile and goes through `_build_kwargs_from_profile` instead. The legacy-path block was therefore dead code. This commit pivots the fix to where it actually fires: - New `DeepSeekProfile` in `plugins/model-providers/deepseek/__init__.py` overrides `build_api_kwargs_extras` to emit DeepSeek's expected wire format (mirrors `KimiProfile`): {"reasoning_effort": "<low|medium|high|max>", "extra_body": {"thinking": {"type": "enabled" | "disabled"}}} - Model gating: only `deepseek-v4-*` and `deepseek-reasoner` emit thinking control. `deepseek-chat` (V3) is untouched — current behavior. - Effort mapping: low/medium/high passthrough, xhigh/max → max, unset → omitted (DeepSeek server applies its own default). - Revert the legacy-path additions from PR #15251 — they were dead code, and the `_copy_reasoning_content_for_api` strip block specifically would have nullified the existing reasoning_content padding machinery (`_needs_deepseek_tool_reasoning` → space-pad on replay) that the active provider already relies on for replay correctness. - Unit tests pin the wire-shape contract and the model gating rules (26 tests, all passing). Existing transport + provider profile suites (321 tests) continue to pass. - AUTHOR_MAP: map twebefy@gmail.com → tw2818 for release notes credit. Closes #15700, #17212, #17825. Co-authored-by: tw2818 <twebefy@gmail.com>
Contributor
🔎 Lint report:
|
| Rule | Count |
|---|---|
invalid-argument-type |
3 |
unresolved-import |
1 |
First entries
run_agent.py:7614: [invalid-argument-type] invalid-argument-type: Argument to function `build_anthropic_client` is incorrect: Expected `str`, found `str | dict[Unknown, Unknown] | Any | ... omitted 3 union elements`
run_agent.py:13897: [invalid-argument-type] invalid-argument-type: Argument to function `_is_oauth_token` is incorrect: Expected `str`, found `str | dict[Unknown, Unknown] | Any | ... omitted 3 union elements`
tests/plugins/model_providers/test_deepseek_profile.py:15: [unresolved-import] unresolved-import: Cannot resolve imported module `pytest`
run_agent.py:13900: [invalid-argument-type] invalid-argument-type: Argument to function `len` is incorrect: Expected `Sized`, found `(str & ~AlwaysFalsy) | (dict[Unknown, Unknown] & ~AlwaysFalsy) | (Any & ~AlwaysFalsy) | ... omitted 3 union elements`
✅ Fixed issues (3):
| Rule | Count |
|---|---|
invalid-argument-type |
3 |
First entries
run_agent.py:13900: [invalid-argument-type] invalid-argument-type: Argument to function `len` is incorrect: Expected `Sized`, found `(str & ~AlwaysFalsy) | (dict[Unknown | str, Unknown | str | dict[str, str]] & ~AlwaysFalsy) | (Any & ~AlwaysFalsy) | ... omitted 3 union elements`
run_agent.py:13897: [invalid-argument-type] invalid-argument-type: Argument to function `_is_oauth_token` is incorrect: Expected `str`, found `str | dict[Unknown | str, Unknown | str | dict[str, str]] | Any | ... omitted 3 union elements`
run_agent.py:7614: [invalid-argument-type] invalid-argument-type: Argument to function `build_anthropic_client` is incorrect: Expected `str`, found `str | dict[Unknown | str, Unknown | str | dict[str, str]] | Any | ... omitted 3 union elements`
Unchanged: 4333 pre-existing issues carried over.
Diagnostics are surfaced as warnings — this check never fails the build.
Collaborator
6 tasks
19 tasks
5 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
DeepSeek V4 /
deepseek-reasonerrequests now go out with the explicitextra_body.thinking+reasoning_effortparameters they require, instead of silently defaulting to thinking-on and 400ing on the next turn.Root cause. DeepSeek has a registered
ProviderProfile, so requests go through_build_kwargs_from_profile. The base profile'sbuild_api_kwargs_extrasreturns({}, {})— so every request toapi.deepseek.comwent out with nothinkingparameter and noreasoning_effort. DeepSeek defaults tothinking=enabled, returnsreasoning_content, then 400s on subsequent turns. Confirmed empirically by instantiating the live profile.@tw2818's PR #15251 correctly diagnosed this but placed the fix in
build_kwargs's legacy fallback path, which DeepSeek never reaches. Cherry-picked their commit to preserve authorship, then pivoted the fix to the profile path in the follow-up commit.Changes
plugins/model-providers/deepseek/__init__.py: newDeepSeekProfile(ProviderProfile)mirroringKimiProfile. Emitsextra_body.thinking = {type: enabled|disabled}+ top-levelreasoning_effort. Model gating: V4 family +deepseek-reasoneronly;deepseek-chat(V3) untouched.scripts/release.py: AUTHOR_MAP entry for @tw2818.tests/plugins/model_providers/test_deepseek_profile.py: 26 unit tests pinning wire shape, model gating, effort mapping, and full-kwargs integration through the transport.Validation
deepseek-v4-prowire format{}(no thinking, no effort){reasoning_effort, extra_body.thinking}tests/run_agent/test_deepseek_v4_thinking_live.py::_thinking_kwargs)Closes #15700.
Closes #17212.
Closes #17825.
Co-authored-by: tw2818 twebefy@gmail.com