Compare

Put models head to head.

Pick up to four models and read them side by side. The best value in every row is called out. The selection lives in the URL, so any comparison is a link you can share.

Claude Fable 5.1new×Claude Opus 5new+Claude Opus 4.8+GPT-5.6 Solnew+Claude Fable 5new+Gemini 3.5 Flash Litenew+GPT-5.5×Claude Sonnet 5new+Muse Spark 1.3 Contributornew+GPT-5.6 Terranew+Muse Spark 1.3new+GPT-6 Astranew+Mistral Medium 3.5new+Mistral Largenew+GPT-5.6 Lunanew+GPT-5.4 Mini+GPT-5.4 Nano+Grok 4.20+Grok 4.3+DeepSeek V3.2+Qwen3.7 Max+Mistral Small+MiniMax M3new+DeepSeek V4 Flashnew+Qwen3.7 Plus+DeepSeek R1+DeepSeek V4 Pronew+Nova 2 Litenew+Ministral 14B+Nemotron 3 Supernew+Llama 4 Mavericknew+GLM-5.2new+Qwen3.7 Flashnew+Llama 4 Scoutnew+Llama 3.3 70B+2/4 selected

Metric	Claude Fable 5.1#1 · Claude Fable 5.1	GPT-5.5#7 · GPT-5.5
Accuracyhigher is better	79.4%	76%
Hallucinationlower is better	2.6%	6.2%
Cost / 1klower is better	$674	$316
Valueacc. pts per $/1k	0.12	0.24
p50 latencylower is better	61.2s	90s
Run-to-run σlower is better	±0	±0
Reliabilityhigher is better	100%	100%

Cost × accuracy

selected in focus