⌕Is ___ SOTA?
benchmarks / fictionlive-120k

Fiction.liveBench 120k

120k token score. Status active. Contamination risk low. Maintainer Fiction.live / Epoch collection. Source.

ModelScoreDateKind
Chutes/llama 4 Maverick 17B 128e Instruct0.3642025-04-06observation
Chutes/qwen3 235B A22b Thinking 25070.6882025-07-25observation
Claude 3.7 Sonnet0.3442025-02-24observation
Claude 3.7 Sonnet 20250219 8K0.5312025-02-24observation
Claude Opus 40.3752025-05-22observation
Claude Sonnet 40.3642025-05-22observation
Deepinfra/qwen3 Next 80B A3b Instruct0.6252025-09-11observation
DeepSeek R10.3332025-01-20observation
Gemini 2.0 Flash 0010.6252025-02-05observation
Gemini 2.0 Flash Thinking Exp 01.210.3752025-01-21observation
Gemini 2.5 Flash Lite Preview 06.17 Thinking0.2192025-06-17observation
Gemini 2.5 Flash Preview 05.200.6882025-05-20observation
Gemini 2.5 Pro Exp 03.250.9062025-03-25observation
Gemini 2.5 Pro Preview 05.060.7192025-05-06observation
Gemini 2.5 Pro Preview 06.050.8752025-06-05observation
GPT-4.10.6252025-04-14observation
GPT-4.5 Preview0.6392025-02-27observation
GPT-50.9692025-08-07observation
GPT-5 Mini0.6252025-08-07observation
Grok 4 07090.9692025-07-09observation
Grok 4 Fast0.752025-09-19observation
Kimi K2 50.7812026-01-27observation
O10.5312024-12-17observation
O312025-04-16observation
O3 Mini0.4382025-01-31observation
O3 Pro0.8892025-06-10observation
O4 Mini0.6252025-04-16observation
ChatGPT 4O 01.29 20250.6562025-01-29observation
DeepSeek V3.10.5312025-08-21observation
Fireworks/kimi K2 Instruct 09050.4062025-09-05observation