Overall score
Best LLM Overall
#1 now: Claude Fable 5.1 · 83.4
See the ranking// rankings
One ranking per use case, each sorted on a single published metric so you can see exactly why a model is on top. The current leader is shown on each card.
Overall score
#1 now: Claude Fable 5.1 · 83.4
See the rankingCoding score
#1 now: Claude Opus 5.5 · 89.3
See the rankingAgentic coding score
#1 now: DeepSeek V4.1 Flash · 77.3
See the rankingReasoning score
#1 now: GPT-6 Astra · 92.7
See the rankingMathematics score
#1 now: Claude Opus 5.5 · 97.1
See the rankingData analysis score
#1 now: GPT-6 Astra · 83.0
See the rankingInstruction following score
#1 now: Gemini 3.8 Flash · 81.4
See the ranking$ per point
#1 now: DeepSeek V4 Flash 0423 · $0.0083
See the rankingBlended / 1M
#1 now: Mistral Nemo · $0.022
See the rankingContext
#1 now: SpaceXAI: Grok 4.20 Multi-Agent · 2M
See the rankingOverall score
#1 now: Claude Fable 5.1 · 83.4
See the rankingOverall score
#1 now: Qwen3.8 27B · 75.3
See the ranking