fix(agent): reset stale token calibration on model switch (#23767) - #50137
Merged
Conversation
ContextCompressor.update_model() recomputed context_length/threshold/budgets but kept the cross-call calibration state (last_real_prompt_tokens, last_rough_tokens_when_real_prompt_fit, last_compression_rough_tokens, awaiting_real_usage_after_compression, _ineffective_compression_count) from the PREVIOUS model. Those fields encode 'the provider proved this prompt fit' / 'preflight can be deferred' decisions valid only for the model that produced them. Carried across a switch to a smaller-context model, should_defer_preflight_to_real_usage() used the old model's 'it fit' history to SKIP a preflight compression the new model actually needed — sending an oversized prompt the provider rejects (#23767). update_model() now clears that state; the new model's first response repopulates it via update_from_response(). Verified E2E: after a 200K->65,536 switch, defer no longer suppresses and should_compress fires on an over-threshold estimate.
Contributor
🔎 Lint report:
|
| Rule | Count |
|---|---|
invalid-assignment |
1 |
First entries
tests/run_agent/test_credits_notices_toggle.py:76: [invalid-assignment] invalid-assignment: Object of type `None` is not assignable to attribute `_credits_session_start_micros` of type `int`
✅ Fixed issues (2):
| Rule | Count |
|---|---|
unresolved-attribute |
2 |
First entries
tests/run_agent/test_credits_notices_toggle.py:76: [unresolved-attribute] unresolved-attribute: Unresolved attribute `_credits_session_start_micros` on type `AIAgent`
run_agent.py:2984: [unresolved-attribute] unresolved-attribute: Object of type `Self@get_credits_spent_micros` has no attribute `_credits_session_start_micros`
Unchanged: 5919 pre-existing issues carried over.
Diagnostics are surfaced as warnings — this check never fails the build.
alt-glitch
added
type/bug
2 tasks
This was referenced Jun 22, 2026
Merged
waefrebeorn
pushed a commit
to waefrebeorn/slermes
that referenced
this pull request
Jul 2, 2026
…ibration-on-model-switch fix(agent): reset stale token calibration on model switch (NousResearch#23767)
habarmc1223-sudo
pushed a commit
to habarmc1223-sudo/hermes-agent-fluxmem
that referenced
this pull request
Jul 8, 2026
…ibration-on-model-switch fix(agent): reset stale token calibration on model switch (NousResearch#23767)
santhreal
pushed a commit
to santhreal/hermes-agent
that referenced
this pull request
Jul 13, 2026
…ibration-on-model-switch fix(agent): reset stale token calibration on model switch (NousResearch#23767)
Gravezzz
pushed a commit
to Gravezzz/hermes-agent
that referenced
this pull request
Jul 21, 2026
…ibration-on-model-switch fix(agent): reset stale token calibration on model switch (NousResearch#23767)
leewenjie
pushed a commit
to leewenjie/hermes-agent
that referenced
this pull request
Aug 7, 2026
…ibration-on-model-switch fix(agent): reset stale token calibration on model switch (NousResearch#23767)
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
2 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
ContextCompressor.update_model()now clears the cross-call token-calibration state captured under the previous model. Without this, after switching an active session onto a smaller-context model, the old model's "this prompt fit / preflight can be deferred" history suppressed a preflight compression the new model actually needed — sending an oversized prompt the provider rejects (part of #23767).Root cause
update_model()recomputedcontext_length/threshold_tokens/ budgets on a switch, but keptlast_real_prompt_tokens,last_rough_tokens_when_real_prompt_fit,last_compression_rough_tokens,awaiting_real_usage_after_compression, and_ineffective_compression_countfrom the old model.should_defer_preflight_to_real_usage()then used the old model's "it fit" signal to skip compression on the new, smaller model.Changes
agent/context_compressor.py:update_model()resets the calibration/usage fields above (andlast_prompt_tokens/last_completion_tokens/last_total_tokens) after recomputing budgets. The new model's first response repopulates them viaupdate_from_response(). Settinglast_prompt_tokens=0is the documented "no data → use rough estimate" state, so the post-response path falls back safely rather than skipping compression.Validation
tests/agent/test_context_compressor.py(+2 new)should_compress(threshold+5k)fires → preflight compresses instead of sending oversizedPart of #23767 (does not close it on its own — sibling fixes tracked separately).
Infographic
Image generation is unavailable in this environment (FAL_KEY unset, no managed-provider credits), so the infographic could not be generated here. To be attached once image generation is available.