BigCodeBench — Results

Rank	Model	Score
1	deepseek-v4-1-flash	60.6
2	ling-3-0-flash-base	52.19
3	ling-3-0-tiny-base	42.89
4	claude-sonnet-3-7	0.358
5	deepseek-r1-0528	0.351
6	deepseek-v3	0.345
7	gemini-2-5-pro	0.331
8	gpt-4-1	0.328
9	deepseek-v3-0324	0.318
10	gpt-4-1-mini	0.318
11	gpt-4o	0.311
12	qwen2-5-coder-32b-instruct	0.308
13	claude-3-5-sonnet	0.304
14	claude-3-5-haiku	0.301
15	claude-3-5-sonnet-v1	0.294
16	gemini-2-0-flash	0.287
17	llama-4-maverick	0.284
18	llama-3-3-70b-instruct	0.284
19	phi-4	0.274
20	llama-3-1-405b-instruct	0.264
21	claude-3-opus	0.26
22	gemma-3-27b	0.26
23	gemini-1-5-pro	0.254
24	qwen2-5-72b-instruct	0.254
25	llama-3-1-70b-instruct	0.254

B

BigCodeBench — Evaluations

1 phaseActive

No evaluations yet for this benchmark.