MistralMistral Small 4 (Reasoning)
- Released
- Mar 16, 2026
- Tracked since
- 2026-07-23
Intelligence index11.3Ranks 346 of 664 models with this metric
Coding index26.6Ranks 172 of 259 models with this metric
Agentic index0.8Ranks 145 of 156 models with this metric
Capability profile
5 axes vs. site averageMistral Small 4 (Reasoning)
Site average
Pricing
per 1M tokens- Input
- $0.150
- Output
- $0.600
- Cache read
- $0.010
Performance
- Output throughput
- 161 tokens/sec
- Time to first token
- 770ms
- First answer token
- 13.23s
- End-to-end response
- 16.34s
History
6Intelligence index
19.6→11.3
Output
Unchanged across 62 days of records
Output throughput
165→161
Cheaper at this level
Models with a comparable intelligence index and a lower output price- Qwen3.5 4B (Reasoning)
AlibabaIntelligence 13.1$0.150−75% - Qwen3.5 4B (Non-reasoning)AlibabaIntelligence 10.8$0.150−75%
- DeepSeek V4 Flash 0420 (Non-reasoning)
DeepSeekIntelligence 18.9$0.200−67% - Qwen3.5 9B (Reasoning)AlibabaIntelligence 13.7$0.200−67%
- Hy3-preview (Reasoning)
TencentIntelligence 22.7$0.210−65%