Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage & billing
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Usage & billing
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksNano RPCEarn crypto
  1. Models
  2. Text
  3. Mistral

Mistral Small 3 24B (2501)

Mistral Small 3 24B (2501) hosted by IONOS in Berlin, Germany. Zero data retention.

  • Tool Calling

Added Sep 10, 2026

Model weights

Pricing

Auto routing · per 1M tokens
Input
$0.116
Output
$0.347
Compare provider prices

Specifications

Context window
32.8K
Max output
8.2K
Avg output (7d)
156 tokens
Longer than 9% of models

Benchmarks

Sourced from LMArena.

Arena Score

1274.4

Overall Rank

#313 / 421

Votes

14,681

Confidence Interval

1268.5 - 1280.3

Category Scores

Coding

#304 / 416

2,083 votes

1311.9

Math

#292 / 404

1,683 votes

1261.5

Longer Query

#311 / 399

1,767 votes

1279.6

Creative Writing

#315 / 419

2,415 votes

1226.6

Instruction Following

#313 / 421

5,485 votes

1255.2

Hard Prompts

#304 / 421

3,569 votes

1284.8

Additional Categories
20

Korean

#231 / 289

231 votes

1239.0

Japanese

#255 / 288

269 votes

1149.7

German

#258 / 320

301 votes

1255.9

French

#262 / 301

142 votes

1252.9

Industry Mathematical

#293 / 401

1,447 votes

1270.3

Industry Medicine And Healthcare

#297 / 390

654 votes

1286.1

Expert

#304 / 371

754 votes

1258.6

Hard Prompts English

#304 / 419

2,355 votes

1294.0

Industry Software And It Services

#305 / 421

3,478 votes

1305.3

Russian

#305 / 385

1,418 votes

1261.6

English

#307 / 421

8,740 votes

1294.3

Industry Life And Physical And Social Science

#311 / 419

2,579 votes

1293.8

Exclude Ties

#313 / 421

9,628 votes

1187.6

Chinese

#314 / 395

1,054 votes

1266.3

Industry Business And Management And Financial Operations

#314 / 414

1,443 votes

1259.1

Multi Turn

#315 / 419

2,151 votes

1249.6

Industry Entertainment And Sports And Media

#316 / 419

2,680 votes

1224.9

Industry Legal And Government

#316 / 393

869 votes

1275.3

Industry Writing And Literature And Language

#316 / 420

3,942 votes

1247.9

Non English

#320 / 421

5,941 votes

1242.7

Published 2026-11-09 · Matched as Mistral Small 24B Instruct (2501)

LMArena Dataset

Providers

Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)

Auto
Auto (ZDR)
$0.05$0.08N/AN/AN/A
Auto
Auto (Anonymized)
$0.05$0.08N/AN/AN/A
DeepInfra
DeepInfra
FP8
US
$0.0525$0.084N/A0.4s69 tps
Kitani
Kitani
INT4
GLB
$0.109$0.343$0.0546N/AN/A
IONOS
IONOS
EU
$0.116$0.347N/AN/AN/A
Privacy:ZDR, prompts not retainedNo training on promptsBest latency and throughput are highlighted.
Batch API
ProviderInputOutputCached input
Auto
Automistralai/mistral-small-24b-instruct-2501
$0.05/M$0.08/MN/A

Batch rows for this model use normal routing and cost the same as standard API requests. Set provider in a row to choose a provider.

Related text models

Compare Mistral Small 3 24B (2501) with similar models from the same provider or model family.

Mistral Small 4 119B Thinking

mistralai/mistral-small-4-119b-2603:thinking

Mistral Small 4 with reasoning enabled (reasoning_effort=high). A hybrid MoE model with deep step-by-step reasoning for complex prompts, coding, and multi-step problem solving.

Mistral Small 4 119B

mistralai/mistral-small-4-119b-2603

Mistral Small 4 is a hybrid MoE model that unifies instruct, reasoning, and coding behavior in a single multimodal model. It supports text and image input, native function calling, JSON output, and per-request reasoning effort controls.

Mistral Devstral Small 2505

mistralai/devstral-small-2505

OpenHands+Devstral is 100% local 100% open, and is SOTA for the category on SWE-Bench Verified: 46.8% accuracy.

Mistral Nemo

mistralai/mistral-nemo-instruct-2407

12B parameter model with multilingual support.

Mixtral 8x22B

mistralai/mixtral-8x22b-instruct-v0.1

Mixtral 8x22B is a powerful sparse Mixture of Experts (MoE) model with 141B total parameters and 39B active per token. Features a 64K context window, exceptional math performance, and cost-efficient inference. Supports English, French, German, Spanish, and Italian. Apache 2.0 licensed.

Devstral 2 123B

mistralai/devstral-2-123b-instruct-2512

Devstral 2 123B is a 123 billion parameter model from Mistral AI optimized for coding and development tasks. Features advanced reasoning capabilities for software engineering workflows.