mirrors/hermes-agent
1
0
Fork 0
mirror of https://github.com/NousResearch/hermes-agent.git synced 2026-07-20 15:33:54 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 7

fix(ollama): emit top-level reasoning_effort=none on /v1/chat/completions (#25758)

Browse source 
Ollama's /v1/chat/completions silently ignores extra_body.think (it only
honours it on /api/chat — ollama/ollama#14820), so agent.reasoning_effort:
none never actually disabled thinking on OpenAI-compatible Ollama routes.
Emit the top-level reasoning_effort='none' field (which Ollama respects)
alongside think=False (kept for proxies and the native /api/chat path).

The PR's second half (propagating reasoning_config to the background-review
fork) already landed on main via agent/background_review.py, so only the
provider-profile change is salvaged here, resolved onto the current
GLM/effort-aware profile.

Salvaged from PR #29820 by @Epoxidex.
This commit is contained in:
Epoxidex 2026-05-21 15:01:14 +03:00 • committed by Teknium
parent a7ef17da7a
commit 8662254ab2
2 changed files with 20 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

14
tests/plugins/model_providers/test_custom_profile.py
View file

@ -49,20 +49,26 @@ class TestCustomReasoningWireShape:
assert tl == {}
def test_disabled_sends_think_false(self, custom_profile):
"""enabled=False → extra_body.think = False (Ollama thinking-off flag)."""
"""enabled=False → reasoning_effort='none' top-level + think=False.
Both fields are required: Ollama's /v1/chat/completions silently
ignores extra_body.think (only /api/chat honours it — ollama#14820)
but respects top-level reasoning_effort (#25758). think=False stays
for proxies and the native /api/chat path.
"""
eb, tl = custom_profile.build_api_kwargs_extras(
reasoning_config={"enabled": False}, model="glm-5.2"
)
assert eb == {"think": False}
assert tl == {}
assert tl == {"reasoning_effort": "none"}
def test_effort_none_sends_think_false(self, custom_profile):
"""effort='none' is the disable alias → think=False, no effort."""
"""effort='none' is the disable alias → same dual emission."""
eb, tl = custom_profile.build_api_kwargs_extras(
reasoning_config={"enabled": True, "effort": "none"}, model="glm-5.2"
)
assert eb == {"think": False}
assert tl == {}
assert tl == {"reasoning_effort": "none"}
@pytest.mark.parametrize(
"effort", ["minimal", "low", "medium", "high", "xhigh", "max"]