Mistral: Mistral Small 3

Text input Text output Free Option
Author's Description

Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it features both pre-trained and instruction-tuned versions designed for efficient local deployment. The model achieves 81% accuracy on the MMLU benchmark and performs competitively with larger models like Llama 3.3 70B and Qwen 32B, while operating at three times the speed on equivalent hardware. [Read the blog post about the model here.](https://mistral.ai/news/mistral-small-3/)

Key Specifications
Cost
$
Context
32K
Parameters
24B
Released
Jan 30, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Stop Presence Penalty Logit Bias Top P Temperature Min P Seed Frequency Penalty Logprobs Max Tokens Top Logprobs
Performance Summary

Mistral Small 3, a 24B-parameter model, demonstrates strong overall performance, particularly excelling in speed and cost-efficiency. It consistently ranks in the top tier for speed (67th percentile) and offers highly competitive pricing (90th percentile), making it an attractive option for cost-sensitive applications. Notably, the model exhibits exceptional reliability with a 100% success rate across all benchmarks, indicating robust and consistent operation. In terms of specific benchmark performance, Mistral Small 3 shows remarkable proficiency in Instruction Following, achieving perfect accuracy in one instance and demonstrating strong capabilities in complex multi-step instructions. It also performs very well in General Knowledge and Ethics, scoring 98% in both categories. While its accuracy in Coding (79%), Email Classification (97%), and Reasoning (56%) is respectable, these areas represent potential avenues for further optimization compared to its standout performance in other domains. The model's ability to compete with larger models like Llama 3.3 70B and Qwen 32B at three times the speed underscores its efficiency and optimized design for low-latency tasks.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.02
Completion $0.08

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Kluster
Kluster | mistralai/mistral-small-24b-instruct-2501 32K $0.02 / 1M tokens $0.08 / 1M tokens
DeepInfra
DeepInfra | mistralai/mistral-small-24b-instruct-2501 32K $0.05 / 1M tokens $0.08 / 1M tokens
Enfer
Enfer | mistralai/mistral-small-24b-instruct-2501 28K $0.02 / 1M tokens $0.08 / 1M tokens
NextBit
NextBit | mistralai/mistral-small-24b-instruct-2501 32K $0.02 / 1M tokens $0.08 / 1M tokens
Mistral
Mistral | mistralai/mistral-small-24b-instruct-2501 32K $0.1 / 1M tokens $0.3 / 1M tokens
Ubicloud
Ubicloud | mistralai/mistral-small-24b-instruct-2501 32K $0.02 / 1M tokens $0.08 / 1M tokens
Together
Together | mistralai/mistral-small-24b-instruct-2501 32K $0.8 / 1M tokens $0.8 / 1M tokens
Enfer
Enfer | mistralai/mistral-small-24b-instruct-2501 28K $0.02 / 1M tokens $0.08 / 1M tokens
Chutes
Chutes | mistralai/mistral-small-24b-instruct-2501 32K $0.02 / 1M tokens $0.08 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Free Executions Accuracy Cost Duration
Other Models by mistralai