Skip to content

fix(agent): disable stale stream timeout for local providers #6368

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
teknium1 merged 1 commit into from
Apr 9, 2026
Merged

fix(agent): disable stale stream timeout for local providers #6368

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
29 changes: 18 additions & 11 deletions run_agent.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4728,18 +4728,25 @@ def _call():
self._close_request_openai_client(request_client, reason="stream_request_complete")

_stream_stale_timeout_base = float(os.getenv("HERMES_STREAM_STALE_TIMEOUT", 180.0))
# Scale the stale timeout for large contexts: slow models (like Opus)
# can legitimately think for minutes before producing the first token
# when the context is large. Without this, the stale detector kills
# healthy connections during the model's thinking phase, producing
# spurious RemoteProtocolError ("peer closed connection").
_est_tokens = sum(len(str(v)) for v in api_kwargs.get("messages", [])) // 4
if _est_tokens > 100_000:
_stream_stale_timeout = max(_stream_stale_timeout_base, 300.0)
elif _est_tokens > 50_000:
_stream_stale_timeout = max(_stream_stale_timeout_base, 240.0)
# Local providers (Ollama, oMLX, llama-cpp) can take 300+ seconds
# for prefill on large contexts. Disable the stale detector unless
# the user explicitly set HERMES_STREAM_STALE_TIMEOUT.
if _stream_stale_timeout_base == 180.0 and self.base_url and is_local_endpoint(self.base_url):
_stream_stale_timeout = float("inf")
logger.debug("Local provider detected (%s) — stale stream timeout disabled", self.base_url)
else:
_stream_stale_timeout = _stream_stale_timeout_base
# Scale the stale timeout for large contexts: slow models (like Opus)
# can legitimately think for minutes before producing the first token
# when the context is large. Without this, the stale detector kills
# healthy connections during the model's thinking phase, producing
# spurious RemoteProtocolError ("peer closed connection").
_est_tokens = sum(len(str(v)) for v in api_kwargs.get("messages", [])) // 4
if _est_tokens > 100_000:
_stream_stale_timeout = max(_stream_stale_timeout_base, 300.0)
elif _est_tokens > 50_000:
_stream_stale_timeout = max(_stream_stale_timeout_base, 240.0)
else:
_stream_stale_timeout = _stream_stale_timeout_base

t = threading.Thread(target=_call, daemon=True)
t.start()
Expand Down
Loading