modelgrep
M

Mistral: Mistral Small 3

mistralai/mistral-small-24b-instruct-2501

Cheaper than 93% of paidJSON
Use via OpenRouter ↗
Intelligence
Design Elo
Speed
47
180th fastest
Latency
270ms
first token
Input price
$0.050
21st cheapest
Context
33K
16K max out

How it compares

Faster than39%
of all ranked models
Cheaper than93%
of all ranked models

Overview

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...

Benchmarks

independent · via OpenRouter
Artificial Analysis
GPQA Diamond
46%
Humanity's Last Exam
4%
SciCode
24%
Tau²-Bench (agentic)
20%

Providers & pricing (1)

ProviderIn $/MOut $/MUptime
DeepInfrafp8$0.050$0.080100%

Specifications

Context window33K
Max output16K
Knowledge cutoffOct 2023
Input modalitiestext
Output modalitiestext
Prompt caching
Cache read price
ModeratedNo

Mistral Small 3 FAQ

How much does Mistral Small 3 cost?

Mistral Small 3 costs $0.050 per million input tokens and $0.080 per million output tokens via OpenRouter, making it 21st cheapest of 298 paid models.

How fast is Mistral Small 3?

Mistral Small 3 generates around 47 tokens per second with 270ms time-to-first-token (p50), the 180th fastest tracked model.

What is Mistral Small 3's context window?

Mistral Small 3 supports a 33K-token context window and can output up to 16K tokens. It accepts text input.

Compare head-to-head