Skip to content

Model Library

MistralReleased January 30, 2025

Mistral Small 3

Mistral · Released January 30, 2025

View public research Browse the library

Intelligence score

7

At a glance

Intelligence
7
Speed
165.6tokens/sec
Latency
770mstime to first token
Input price
$0.10per 1M tokens
Output price
$0.30per 1M tokens
Context window
32Ktokens
Parameters
24billions

Summary

Comparison summary

Mistral Small 3 scores 7 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 6). Mistral Small 3 generates output at 165.6 tokens per second (based on Mistral's API), which is well above average compared to other open weight non-reasoning models of similar size (median: 101.8 t/s). Mistral Small 3 costs $0.10 per 1M input tokens (better than average, median: $0.15) and $0.30 per 1M output tokens (better than average, median: $0.32), based on Mistral's API.

Specifications

Technical specifications

Model nameMistral Small 3
ProviderMistral
Release dateJanuary 30, 2025
Intelligence score7
Speed165.6 tokens/sec
Latency (TTFT)0.77s
Input price$0.10 / 1M tokens
Output price$0.30 / 1M tokens
Context window32,000 tokens
Parameters24B
ReasoningNo
Input modalitiestext
Output modalities—
Open weightsYes
LicenseApache 2
API availability—
SourceModel Intelligence legacy profile: mistral-small-3

Pricing

Token prices

Price per 1M tokens

Mistral Small 3, as listed in the catalog record

Input $0.10 and output $0.30 per 1M tokens. Missing prices are hatched and marked "Not published".
  1. Input$0.10
  2. Output$0.30

Model Intelligence legacy profile: mistral-small-3

FAQ · 19

Frequently asked questions

Mistral Small 3 was released on January 30, 2025.

Mistral Small 3 was created by Mistral.

Mistral Small 3 scores 7 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 6).

Mistral Small 3 generates output at 165.6 tokens per second (based on Mistral's API), which is well above average compared to other open weight non-reasoning models of similar size (median: 101.8 t/s).

Mistral Small 3 has a time to first token (TTFT) of 0.77s (based on Mistral's API), which is very competitive compared to other open weight non-reasoning models of similar size (median: 1.54s).

Mistral Small 3 costs $0.10 per 1M input tokens (better than average, median: $0.15) and $0.30 per 1M output tokens (better than average, median: $0.32), based on Mistral's API.

Mistral Small 3 costs $0.10 per 1M input tokens and $0.30 per 1M output tokens (based on Mistral's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.12 per 1M tokens. Pricing may vary by provider.

No, Mistral Small 3 is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.

Mistral Small 3 supports text input.

Mistral Small 3 supports text output.

No, Mistral Small 3 does not support image input. It can only process text.

No, Mistral Small 3 is not multimodal. It only supports text input.

Mistral Small 3 has a context window of 32k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Mistral Small 3 is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Mistral Small 3 has 24 billion parameters.

Mistral Small 3 is released under the Apache 2.0 license. This license allows commercial use.

Mistral Small 3 achieves a score of 7 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, Mistral Small 3 is available via API through 2 providers.

Mistral Small 3 is available through 2 API providers.