Calibrate reasoning: add reasoningFrequency, update defaults#124
Merged
Conversation
…aults Calibrate reasoning output based on analysis of 127 Models Agent conversations: - reasoningOutputSize: 500 → 265 (calibrated mean) - New reasoningFrequency parameter (default 0.47): fraction of turns with reasoning output. Uses Bresenham-style distribution for deterministic, even spacing across cycles. Previously the sim charged reasoning on every turn at 500 tokens — a ~3-4x overstatement vs real data (only 47% of turns include thinking, at 265 avg). Closes #94 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
reasoningFrequencyparameter (0.0–1.0, default 0.47) — fraction of assistant turns that include reasoning outputreasoningOutputSizedefault from 500 → 265 tokens (mean from 127 Models Agent conversations)Engine Change
The sim previously overcharged reasoning output by ~3-4x (size overstatement × frequency overstatement). Since reasoning is billed at output price ($25/M), this inflates all absolute cost numbers. Rankings are unaffected (affects all strategies equally), but absolute costs were misleading.
Prior findings impact: All absolute cost numbers from prior experiments are overstated. Strategy rankings and relative comparisons remain valid since reasoning affects all strategies equally. FINDINGS.md will be updated after merge.
Test plan
Closes #94
🤖 Generated with Claude Code