Mistral: Models Intelligence, Performance & Price
Analysis of Mistral's models across key metrics including quality, price, output speed, latency, context window & more. This analysis is intended to support you in choosing the best model provided by Mistral for your use-case.
Most Intelligent
Magistral Medium 1.2Mistral Large 3Devstral 2Mistral Medium 3.1Devstral Small 2Intelligence index
Total 22 models
Fastest
Ministral 3 3BDevstral SmallDevstral Small 2Ministral 3 8BMistral 7BOutput speed
Total 22 models
Lowest Price
Ministral 3 3BDevstral SmallMistral Small 3.2Ministral 3 8BMistral Small 3.1Blended price (per 1M tokens)
Total 22 models
Mistral offers 22 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.
- For intelligence, the top models on Mistral are Magistral Medium 1.2 (27), Mistral Large 3 (23), Devstral 2 (22).
- For output speed, the fastest models are Ministral 3 3B (269 t/s), Devstral Small (197 t/s), Devstral Small 2 (186 t/s). Speed varies significantly across models, with a 73% difference between the fastest and slowest.
- For latency, Ministral 3 3B (0.41s), Mistral 7B (0.52s), Ministral 3 8B (0.54s) offer the lowest time to first token.
- For pricing, Ministral 3 3B ($0.10), Devstral Small ($0.15), Mistral Small 3.2 ($0.15) offer the lowest blended prices per 1M tokens.
- For context window size, Mistral Large 3 (256k), Devstral 2 (256k), Devstral Small 2 (256k) support the largest context windows on Mistral.
- Ministral 3 3B offers both the fastest output and best pricing, making it attractive for throughput-sensitive and cost-conscious applications. Magistral Medium 1.2 leads in intelligence for tasks that require the highest quality.
Intelligence Evaluations
Artificial Analysis Intelligence Index
Intelligence Evaluations
Intelligence vs. Price
Context Window
Context Window
JSON Mode & Function Calling
Function (Tool) Calling & JSON Mode
| Models | Function calling | JSON Mode |
|---|---|---|
 Magistral Medium 1.2, Mistral | ||
| Mistral Large 3, Mistral | ||
| Devstral 2, Mistral | ||
| Mistral Medium 3.1, Mistral | ||
| Devstral Small 2, Mistral | ||
| Mistral Medium 3, Mistral | ||
| Devstral Medium, Mistral | ||
| Mistral Small 4, Mistral | ||
| Magistral Small 1.2, Mistral | ||
| Ministral 3 14B, Mistral | ||
| Devstral Small, Mistral | ||
| Mistral Large 2 (Nov), Mistral |
Pricing
Intelligence vs. Price
Performance Summary
Output Speed vs. Price
Speed
Measured by Output Speed (tokens per second)
Output Speed
Latency
Measured by Time (seconds) to First Token
Time to First Answer Token
End-to-End Response Time
Seconds to output 500 Tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed
End-to-End Response Time vs. Price
Key definitions
Frequently Asked Questions
Common questions about Mistral
Mistral offers 22 models that we track: Magistral Medium 1.2, Mistral Large 3, Devstral 2, Mistral Medium 3.1, Devstral Small 2, Mistral Medium 3, Devstral Medium, Mistral Small 4, Magistral Small 1.2, Ministral 3 14B, Devstral Small, Mistral Large 2 (Nov), Mistral Small 3.2, Ministral 3 8B, Mistral Small 3.1, Ministral 3 3B, Pixtral Large, Mistral Small 3, Mistral Small (Sep), Mistral Small (Feb), Mistral 7B, and Mistral Medium.
The most intelligent model available on Mistral is Magistral Medium 1.2 with an Intelligence Index score of 27.
The fastest model on Mistral by output speed is Ministral 3 3B at 269.2 tokens per second.
The model with the lowest time to first token on Mistral is Ministral 3 3B at 0.41s. Lower latency means faster initial response time.
The most affordable model on Mistral by blended price is Ministral 3 3B at $0.10 per 1M tokens (3:1 input to output ratio).
Prices on Mistral vary up to 41x across models, from $0.10 per 1M tokens for Ministral 3 3B to $4.09 per 1M tokens for Mistral Medium.
Yes, Mistral offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.
Yes, all 22 models on Mistral support JSON mode for structured output.
21 of 22 models on Mistral support function calling (tool use).
Yes, Mistral offers 2 reasoning models: Magistral Medium 1.2 and Magistral Small 1.2. Reasoning models use extended thinking to work through complex problems before providing an answer.
Yes, 16 of 22 models on Mistral are open weight models: Mistral Large 3, Devstral 2, Devstral Small 2, Mistral Small 4, Magistral Small 1.2, Ministral 3 14B, Devstral Small, Mistral Large 2 (Nov), Mistral Small 3.2, Ministral 3 8B, Mistral Small 3.1, Ministral 3 3B, Pixtral Large, Mistral Small 3, Mistral Small (Sep), and Mistral 7B.
Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.
When choosing a model on Mistral, consider: intelligence (for quality-sensitive tasks), output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and features like context window size, JSON mode, or function calling support.