Loading model…
qwen/qwen3.7-plus14/28 in v2 — Resilience, score 69.89%, pass rate 47.06% (216/459)0.98s, output speed 55.07 tok/s$0.35, prompt tokens 31989, completion tokens 2146948/9 passed, low TTFT (1.14s), fast output (55.68 tok/s).29/30, 55/60) with minimal spend.54/60 passed, solid score (250/261).2/60 passed (298/541 score).8/60 passed despite high max score (283/512).11/60 passed (368/541).openai/gpt-5.5 (score 77.90%, pass rate 61.87%), openai/gpt-5.3-codex (77.66%, 61.22%), and qwen/qwen3.7-max (75.27%, 54.90%).qwen/qwen3.7-plus sits mid-pack (14/28) in Resilience, with clear strengths in UI, cost, and speed tasks but significant gaps in refactoring, security, and reasoning. Suitable for lightweight coding and cost-sensitive deployments, but risky for complex refactoring or security-critical workloads.
Overall score % from merged run_models rows (chronological). Only runs that include this model appear as points.
Score % vs pass rate % per category. With 0/1 scorers, both usually line up; with proportional tests, score % reflects partial credit while pass rate counts tests that clear the fixture threshold.
Total estimated spend per scope for this model (bars, left axis) and mean spend per merged result row (line, right axis: total ÷ tests).
Pass rate % per difficulty level — complements the score % view above.
Normalized 0–100 within this model: TTFT (shorter → higher spoke) and decode tok/s (higher → higher spoke). Values come from streamed BLXBench runs merged into overall_ranking.json.
Pass rate % per category for this model (distinct from score %, which reflects partial credit).