APEX/BrowserUse_and_ComputerUse_skills
5
0
Fork 0
Code Issues 6 Pull requests 1 Projects 1 Releases Packages Wiki Activity Actions

Merge pull request #2215 from NousResearch/hermes/hermes-31d7db3b

Browse source 
fix: infer provider from base URL for models.dev context length lookup
This commit is contained in:
Teknium 2026-03-20 12:52:07 -07:00 • committed by GitHub
commit 8e884fb3f1
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 66 additions and 18 deletions
Download patch file Download diff file Expand all files Collapse all files

10
cli.py
View file

@ -1504,7 +1504,7 @@ class HermesCLI:
_cprint(f"{_DIM}└{'─' * (w - 2)}┘{_RST}")
self._reasoning_box_opened = False
def _stream_delta(self, text: str) -> None:
def _stream_delta(self, text) -> None:
"""Line-buffered streaming callback for real-time token rendering.
Receives text deltas from the agent as tokens arrive. Buffers
@ -1514,7 +1514,15 @@ class HermesCLI:
Reasoning/thinking blocks (<REASONING_SCRATCHPAD>, <think>, etc.)
are suppressed during streaming since they'd display raw XML tags.
The agent strips them from the final response anyway.
A ``None`` value signals an intermediate turn boundary (tools are
about to execute). Flushes any open boxes and resets state so
tool feed lines render cleanly between turns.
"""
if text is None:
self._flush_stream()
self._reset_stream_state()
return
if not text:
return