IFBench — Results

Rank	Model	Score
1	qwen3-8-max	82.8
2	nemotron-3-ultra-550b-a55b	81.4
3	qwen3-8-flash-next	81.3
4	hermes-3-70b	81.2
5	nova-2-pro	80.2
6	inkling	79.8
7	qwen3-8-27b	79.5
8	granite-4-2-8b	79.33
9	granite-4-2-30b	77.17
10	gemini-3-1-pro	77.1
11	muse-glimmer	77
12	deepseek-v4-pro	76.5
13	qwen3-5-27b	76.5
14	qwen3-5-397b-a17b	76.5
15	qwen3-5-122b-a10b	76.1
16	kimi-k2-6	76
17	granite-4-2-3b	74.33
18	qwen3-6-plus	74.2
19	glm-5-2	73.3
20	nemotron-3-5-lightning-30b-a3b	72.9
21	gpt-5-6-sol	72.7
22	nemotron-3-super-120b-a12b	72.6
23	kimi-k2-5	70.2
24	qwen3-5-35b-a3b	70.2
25	minicpm5-2b	66.3

IFBench — Evaluations

1 phaseActive

No evaluations yet for this benchmark.