Share your thoughts, 1 month free Claude Pro on usSee more

Long-context Question Answering on FRAMES

73.54Avg@4 Score

GPT5-Nano

Updated 1mo ago

Evaluation Results

Method	Links
GPT5-Nano 2026.01		73.54
GPT-OSS-120B 2026.01		72.69
LONGPAS 2026.01		71.84
Qwen3-30B-A3B-Thinking 2026.01		69.66
RLVR 2026.01		69.66
LONGPAS 2026.01		68.93
RLVR 2026.01		66.26
Gemini-2.5-Flash-Thinking 2026.01		65.78
LONGPAS 2026.01		64.9
LONGPAS 2026.01		64.75
GPT-OSS-20B 2026.01		64.44
Qwen3-30B-A3B-Instruct 2026.01		62.96
RLVR 2026.01		62.74
RLVR 2026.01		60.92
Qwen3-4B-Thinking 2026.01		60.84
LONGPAS 2026.01		60.62
RLVR 2026.01		55.98
LONGPAS 2026.01		55.76
RLVR 2026.01		50.97
Qwen3-4B-Instruct 2026.01		46.81
Qwen2.5-7B-Instruct 2026.01		45.08
LLaMA3.1-8B-Instruct 2026.01		41.84