NOUMENOpen app
NOUMEN

Mistral · Open weights

Mistral Small 3 24B (2501)

Added Sep 10, 2026Open weights

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

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

  • Tools
Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.12
Output
$0.35

Reference prices as published by the model provider. API example

Specifications

Context window
32.8K
Max output
8.2K
Average output (7 days)
403 tokens
Longer than 40% of 468 models

Side by side

Mistral Small 3 24B (2501) compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Mistral Small 3 24B (2501)MistralThis modelGLM 4.5VZ.aiRemove
ProviderMistralZ.ai
Discovery tierOpen weightsOpen weights
AddedSep 10, 2026Nov 22, 2025
Limits
Context window32.77K65.54KBest
Max output8.19K16.38KBest
Average output403 tokens732 tokens
Pricing, USD per 1M tokens
Input$0.12Best$0.60
Output$0.35Best$1.80
Benchmark indexes
Intelligence Index6.76.7
Math Index4.315.3Best
Capabilities
ReasoningNoYes
ToolsYesNo
VisionNoYes
Reasoning effort——

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • Mistral Small 3 24B (2501)
  • GLM 4.5V

Intelligence Index

  • Mistral Small 3 24B (2501)6.7
  • GLM 4.5V6.7

Mistral Small 3 24B (2501): Better than 6% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Math Index

    Composite of the math evaluations

    • Mistral Small 3 24B (2501): 4.3
    • GLM 4.5V: 15.3Top

Document reasoning

  • AA-LCRv1.1

    Reasoning across long documents

    • Mistral Small 3 24B (2501): 0.0%Top
    • GLM 4.5V: 0.0%Top

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Mistral Small 3 24B (2501): 3.8%Top
    • GLM 4.5V: 3.5%

  • GPQA Diamond

    Graduate-level science questions

    • Mistral Small 3 24B (2501): 46.2%
    • GLM 4.5V: 57.3%Top

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • Mistral Small 3 24B (2501): 65.2%
    • GLM 4.5V: 75.1%Top

  • IFBench

    Following precise output instructions

    • Mistral Small 3 24B (2501): 26.4%
    • GLM 4.5V: 28.6%Top

Coding

  • LiveCodeBench

    Recent competitive programming problems

    • Mistral Small 3 24B (2501): 25.2%
    • GLM 4.5V: 35.2%Top

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • Mistral Small 3 24B (2501): 4.3%
    • GLM 4.5V: 15.3%Top

  • AIME

    Competition mathematics

    • Mistral Small 3 24B (2501): 8.0%
    • GLM 4.5V: —

  • MATH-500

    Competition math problems

    • Mistral Small 3 24B (2501): 71.5%
    • GLM 4.5V: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Mistral Small 3 24B (2501): —
    • GLM 4.5V: 6.8%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Mistral Small 3 24B (2501): 19.6%Top
    • GLM 4.5V: 19.6%Top

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Mistral Small 3 24B (2501): 0.0%Top
    • GLM 4.5V: 0.0%Top

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mistralai/mistral-small-24b-instruct-2501",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="mistralai/mistral-small-24b-instruct-2501",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Mistral

Other Mistral models, starting with the same family.