- Intelligence
- 11.3
- Rank 392 of 744 measured models
- Coding
- 26.6
- Rank 192 of 298 measured models
- Output speed
- 165 t/s
- Rank 66 of 403 measured models
- Price/1M
- $0.262
- Rank 192 of 659 measured models
- Context window
- 256K
- Rank 518 of 1199 measured models
- Cost per task
- $0.015
- Rank 17 of 193 measured models
Mistral Small 4 (Reasoning) scores 11.3 on the Artificial Analysis Intelligence Index, ranking 392nd of 744 measured models. The closest models on this index are DeepSeek R1 (Jan '25) (11.4), Grok 4.1 Fast (11.3), and Mistral Small 4 (batch) (11.3). At $0.262 per million tokens (3:1 input/output blended price), it is cheaper than DeepSeek R1 (Jan '25) ($2.50), more expensive than Mistral Small 4 (batch) ($0.131). With an output speed of 165 t/s, it is faster than DeepSeek R1 (Jan '25) (150 t/s).
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- DeepSeek R1 (Jan '25)
DeepSeek · $2.50 · 150 t/s
11.4+0.1$2.50150 t/sCompare - HyperCLOVA X SEED Think (32B)
Naver · — · —
11.4+0.1——Compare - Grok 4.1 Fast
SpaceXSI · — · —
11.3——Compare - Mistral Small 4 (batch)
Mistral · $0.131 · —
11.3$0.131—Compare
