- Intelligence
- 15.3
- Rank 308 of 747 measured models
- Output speed
- 50.7 t/s
- Rank 335 of 405 measured models
- Price/1M
- $1.08
- Rank 391 of 664 measured models
- Context window
- 256K
- Rank 521 of 1207 measured models
Kimi K2 0905 scores 15.3 on the Artificial Analysis Intelligence Index, ranking 308th of 747 measured models. The closest models on this index are Claude 4.5 Haiku (15.4), Claude 3.7 Sonnet (15.3), and o1 (15.2). At $1.08 per million tokens (3:1 input/output blended price), it is cheaper than Claude 4.5 Haiku ($2), Claude 3.7 Sonnet ($6), and o1 ($26.25). With an output speed of 50.7 t/s, it is slower than Claude 4.5 Haiku (88.5 t/s).
Additional benchmarks2
Design Arena · Codecategories48.5%
Design Arena · Website48.3%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Gemini 3.1 Flash-Lite
Google · $0.563 · 385 t/s
15.6+0.3$0.563385 t/sCompare - Claude 4.5 Haiku
Anthropic · $2 · 88.5 t/s
15.4+0.1$288.5 t/sCompare - Claude 3.7 Sonnet
Anthropic · $6 · —
15.3$6—Compare - o1
OpenAI · $26.25 · —
15.2−0.1$26.25—Compare
