live
weekly refresh
basedagi.org
benchmark evidence

BigCodeBench Instruct

BigCodeBench bigcodebench-instruct — practical Python coding benchmark.

winner on BigCodeBench Instruct
direct benchmark result, not a broad vertical composite | source row dated 2000-01-01
scored on 2000-01-01 · stale source data (9646d)
latest mapped results | top 20
#ModelScoreEvidenceTested
1Meta: Llama 4 Maverick
Meta Llama
49.7
model-only
independent_benchmark
2000-01-01
2Qwen2.5 Coder 32B Instruct
Qwen
49.0
model-only
independent_benchmark
2000-01-01
3Meta: Llama 3.3 70B Instruct
Meta Llama
46.9
model-only
independent_benchmark
2000-01-01
4Mistral: Mistral Small 4
Mistralai
36.1
model-only
independent_benchmark
2000-01-01
what this result means

BigCodeBench bigcodebench-instruct — practical Python coding benchmark.

This benchmark contributes direct public evidence. Read its scope before generalizing the result.

A win here is a win on BigCodeBench Instruct. Broad task pages require independent corroboration before naming a general winner.

source record
category: coding
metric: pass_at_1
matched models: 4
latest source date: 2000-01-01
direction: higher is better
inspect upstream source ->