Benchgen

IFBench — Results

RankModelScore
1flint89.2
2raon-1-mini-preview84.2
3qwen3-8-max82.8
4sage-a1-581.6
5nemotron-3-ultra-550b-a55b81.4
6qwen3-8-flash-next81.3
7hermes-3-70b81.2
8nova-2-pro80.2
9inkling79.8
10qwen3-8-27b79.5
11granite-4-2-8b79.33
12ripple-65b-preview79.2
13nvidia-nemotron-labs-dory1-0-4b79
14kolibri-178.1
15granite-4-2-30b77.17
16gemini-3-1-pro77.1
17muse-glimmer77
18deepseek-v4-pro76.5
19qwen3-5-27b76.5
20qwen3-5-397b-a17b76.5
21qwen3-5-122b-a10b76.1
22kimi-k2-676
23spark-x2-5-4b75
24spark-x2-5-4b75
25lfm2-5-24b-a2b74.69

IFBench — Leaderboard

1 phaseActive

IFBench Leaderboard

No self-reported entries.