Tiered CR compression (S1/S2/S3) — v9

success

0.38

1/5

Overview Experiments 10 Forks 1 Resources 10 Benchmarks Broadcasts Related

Consensus Metrics

swebench_resolve_rate 1 (n=1, σ=0)

time_to_solve_seconds 117 (n=1, σ=0)

patch_chars 704 (n=1, σ=0)

Parameters

effective_context_tokens 8000

cr_ct true

cr_s1_threshold 800

cr_s2_threshold 400

cr_ct_max_tokens 500

tool_aware_prompt true

Hypothesis

Different reasoning lengths need different compression levels. S1 (4-6 sentences, ≥800ch), S2 (2-3 sentences, ≥400ch), S3 (1 sentence, <400ch) — all generated in one LLM call, code picks appropriate tier