Anthropic
Claude Sonnet 5
Closed weights
- Intelligence
- 23.2
- Rank 199 of 744 measured models
- Coding
- 66.4
- Rank 66 of 298 measured models
- Output speed
- 58.3 t/s
- Rank 285 of 403 measured models
- Price/1M
- $4
- Rank 532 of 659 measured models
- Context window
- 1M
- Rank 84 of 1199 measured models
Claude Sonnet 5 (Non-reasoning, High) scores 23.2 on the Artificial Analysis Intelligence Index, ranking 199th of 744 measured models. The closest models on this index are GLM 5V Turbo (Reasoning) (23.5), Kimi K2.5 (23.5), and GPT-5 (23.0). At $4 per million tokens (3:1 input/output blended price), it is more expensive than Kimi K2.5 ($1.14) and GPT-5 ($3.44). With an output speed of 58.3 t/s, it is about as fast as Kimi K2.5 (60 t/s), slower than GPT-5 (102 t/s).
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- GLM 5V Turbo (Reasoning)
Z.AI · — · —
23.5+0.3——Compare - Kimi K2.5
Moonshot AI · $1.14 · 60 t/s
23.5+0.3$1.1460 t/sCompare - GPT-5
OpenAI · $3.44 · 102 t/s
23.0−0.2$3.44102 t/sCompare - Nemotron 3 Ultra 550B A55B
NVIDIA · $1.08 · 156 t/s
22.9−0.3$1.08156 t/sCompare
