Mistral Small
Mistral AI
Mistral AI's optimized model balancing performance and cost-effectiveness. Designed for latency-sensitive applications requiring reliable performance without flagship pricing. Outperforms Mixtral 8x7B while offering lower latency. Features 32K context window (some sources indicate 131K support) and 4K max output. Supports function calling and JSON formatting for structured outputs. Excellent for data analysis, research tasks, and everyday AI applications. Positioned as refined intermediary between open-weights and flagship tiers.
Strengths
- Excellent price-to-performance ratio ($1 input vs competitors' $2.50+)
- Optimized for low latency - ideal for real-time applications
- Outperforms Mixtral 8x7B while maintaining efficiency
- Function calling and JSON formatting for structured workflows
- Strong at data analysis and research tasks
- Lower cost than Mistral Large while maintaining quality
- RAG-enabled with same innovations as Mistral Large
Caveats
- 32K context window smaller than competitors (GPT-4o: 128K, Gemini 1.5 Pro: 2M)
- No vision or audio capabilities (text-only)
- Lower ceiling than Mistral Large for complex reasoning
- Being superseded by Mistral Small (Sep '24) - newer version available
- 4K max output tokens lower than flagship models
- Vendor
- Mistral AI
- Context Window
- 32,768 tokens
- Max Output
- 4,096 tokens
- License
- Proprietary
- Release Date
- 2024-02-01
- Modalities
- text
| Benchmark | Score | Max |
|---|---|---|
| livebench_instruction_following | 68.0 | 100.0 |
| livebench_language | 31.8 | 100.0 |
| livebench_coding | 36.2 | 100.0 |
Capabilities
Vision
Audio
Video
Tool Use
Resources
Pricing
$1.00 per 1M input tokens, $3.00 per 1M output tokens