List your company

Instruction Following on AgentIF

66.1CSRClaude-Opus-4.7
41.86848.15954.4560.741May 8, 2026

Evaluation results

MethodLinks
Claude-Opus-4.7Stage=-2026.0566.1
R1-0528-Qwen3-8BStage=Iter32026.0562.6
R1-0528-Qwen3-8BStage=Iter22026.0562.3
R1-0528-Qwen3-8BStage=Iter12026.0561.2
Qwen2.5-7B-InstructStage=Iter32026.0560.5
Qwen2.5-7B-InstructStage=Iter22026.0559.8
Distill-Qwen-14BStage=Iter12026.0559.7
Qwen2.5-7B-InstructStage=Iter12026.0559.3
Distill-Qwen-14BStage=Iter22026.0558.9
GPT-4oStage=-2026.0558.5
Distill-Qwen-14BStage=Iter32026.0558.2
QwQ-32BStage=-2026.0558.1
Llama-3.1-8B-InstructStage=Iter32026.0557.4
R1-0528-Qwen3-8BStage=BASE2026.0557.4
Self-Supervised-7BStage=-2026.0556.7
VERIF-8BStage=-2026.0556.6
Llama-3.1-8B-InstructStage=Iter12026.0555.8
Llama-3.1-8B-InstructStage=Iter22026.0555.2
Distill-Qwen-14BStage=BASE2026.0554.5
Qwen2.5-7B-InstructStage=BASE2026.0554.2
SPAR-8B-DPOStage=-2026.0553.6
Llama-3.1-8B-InstructStage=BASE2026.0553.4
RAIF-7BStage=-2026.0551.9
Qwen2.5-1.5B-InstructStage=Iter32026.0547.5
Crab-7B-DPOStage=-2026.0547.2
Qwen2.5-1.5B-InstructStage=Iter12026.0546.2
Qwen2.5-1.5B-InstructStage=Iter22026.0545.6
Conifer-7B-DPOStage=-2026.0544.3
Qwen2.5-1.5B-InstructStage=BASE2026.0542.8