mirrors/hermes-agent
1
0
Fork 0
mirror of https://github.com/NousResearch/hermes-agent.git synced 2026-04-25 00:51:20 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix(compression): exclude completion tokens from compression trigger (#12026)

Browse source 
Cherry-picked from PR #12481 by @Sanjays2402.

Reasoning models (GLM-5.1, QwQ, DeepSeek R1) inflate completion_tokens
with internal thinking tokens. The compression trigger summed
prompt_tokens + completion_tokens, causing premature compression at ~42%
actual context usage instead of the configured 50% threshold.

Now uses only prompt_tokens — completion tokens don't consume context
window space for the next API call.

- 3 new regression tests
- Added AUTHOR_MAP entry for @Sanjays2402

Closes #12026
This commit is contained in:
Sanjays2402 2026-04-20 05:06:04 -07:00 • committed by Teknium
parent 42c30985c7
commit 570f8bab8f
3 changed files with 68 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

10
run_agent.py
View file

@ -11736,10 +11736,12 @@ class AIAgent:
# should_compress(0) never fires. (#2153)
_compressor = self.context_compressor
if _compressor.last_prompt_tokens > 0:
_real_tokens = (
_compressor.last_prompt_tokens
+ _compressor.last_completion_tokens
)
# Only use prompt_tokens — completion/reasoning
# tokens don't consume context window space.
# Thinking models (GLM-5.1, QwQ, DeepSeek R1)
# inflate completion_tokens with reasoning,
# causing premature compression. (#12026)
_real_tokens = _compressor.last_prompt_tokens
else:
_real_tokens = estimate_messages_tokens_rough(messages)