feat(arcee): Trinity Large Thinking temperature + compression overrides#20473
Merged
Conversation
…verrides Salvage follow-up for PR #20344: - AUTHOR_MAP entry for rob-maron (required by CI) - 17 parametrized tests covering _is_arcee_trinity_thinking, _fixed_temperature_for_model Trinity override, and _compression_threshold_for_model, including sibling-model negatives (trinity-large-preview, trinity-mini) and the OpenRouter slug form.
1 task
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Salvage of #20344 by @rob-maron onto current main.
Summary
Trinity Large Thinking (Arcee reasoning model) now pins temperature to 0.5 and bumps the compression threshold to 0.75, so more raw reasoning context survives before summarization kicks in.
Changes
agent/auxiliary_client.py:_is_arcee_trinity_thinking(), temperature override wired into existing_fixed_temperature_for_model()(auto-fans out to transports, trajectory compressor, mini_swe_runner, and run_agent init — no new wiring needed). New_compression_threshold_for_model()helper.run_agent.py: 6-line plug-in right aftercompression.thresholdis read from config, wrapped in try/except so import failures stay non-fatal.scripts/release.py: AUTHOR_MAP entry for rob-maron.tests/agent/test_arcee_trinity_overrides.py: 17 parametrized tests — bare + OpenRouter-slugged + case/whitespace variants, plus sibling-model negatives (trinity-large-preview,trinity-mini) to confirm overrides don't leak.Validation
scripts/run_tests.sh tests/agent/test_arcee_trinity_overrides.py→ 17 passed.Closes #20344. Original commit authored by @rob-maron preserved via rebase-merge.