cohere/north-mini-code:free66.3%, pass rate 43.4% (199/459)1.16s, output speed 210.9 tok/s$0.00, prompt tokens 25,949, completion tokens 190,64132/60 passed, score 266/360, competitive on factual reliability.31/60 (438/601), showing useful real-world coding assistance.14/60 (73/261), the weakest core category and a major deployment risk.11/60, 360/541) and security (11/60, 279/512) both underperform, limiting high-stakes use.16/60, 374/541) trails, suggesting weak multi-step code transformation.66.3%, 43.4% pass), it sits well below top-3 leaders like openai/gpt-5.5 (77.9%, 61.9% pass) and qwen/qwen3.7-max (75.3%, 54.9% pass), with a large gap in both score and pass rate.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).