NOUMENOpen app
NOUMEN

Mistral · Balanced

Mistral Medium 3

Added Sep 25, 2025

Compare models

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

Loading models…

Use in the app

Mistral Medium 3 delivers frontier performance while being an order of magnitude less expensive. For instance, the model performs at or above 90% of Claude Sonnet 3.7 on benchmarks across the board at a significantly lower cost. On performance, Mistral Medium 3 also surpasses leading open models such as Llama 4 Maverick and enterprise models such as Cohere Command A. On pricing, the model beats cost leaders such as DeepSeek v3, both in API and self-deployed systems.

  • Vision
Input
Text, Images
Output
Text
Discovery tier
Balanced

Pricing

USD per 1M tokens
Input
$0.40
Output
$2.00

Reference prices as published by the model provider. API example

Specifications

Context window
131.1K
Max output
32.8K
Average output (7 days)
683 tokens
Longer than 61% of 468 models

Side by side

Mistral Medium 3 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 Medium 3MistralThis modelMistral Large 4MistralRemove
ProviderMistralMistral
Discovery tierBalancedOpen weights
AddedSep 25, 2025Oct 6, 2026
Limits
Context window131.07K524.29KBest
Max output32.77K—
Average output683 tokens—
Pricing, USD per 1M tokens
Input$0.40Best$0.68
Output$2.00Best$2.09
Cache read—$0.07
Performance
Throughput105.1 tok/s—
Time to first token2.9 s—
Request success (24h)100.0%Best69.0%
Benchmark indexes
Intelligence Index9.0—
Math Index30.3—
Capabilities
ReasoningNoYes
ToolsNoYes
VisionYesYes
Structured outputNoYes
Reasoning effort—none, high

Public evaluations

Benchmarks

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

  • Mistral Medium 3
  • Mistral Large 4

Intelligence Index

  • Mistral Medium 39.0
  • Mistral Large 4—

Mistral Medium 3: Better than 16% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Math Index

    Composite of the math evaluations

    • Mistral Medium 3: 30.3
    • Mistral Large 4: —

Document reasoning

  • AA-LCRv1.1

    Reasoning across long documents

    • Mistral Medium 3: 31.3%
    • Mistral Large 4: —

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Mistral Medium 3: 4.1%
    • Mistral Large 4: —

  • GPQA Diamond

    Graduate-level science questions

    • Mistral Medium 3: 57.8%
    • Mistral Large 4: —

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • Mistral Medium 3: 76.0%
    • Mistral Large 4: —

  • IFBench

    Following precise output instructions

    • Mistral Medium 3: 39.3%
    • Mistral Large 4: —

Coding

  • LiveCodeBench

    Recent competitive programming problems

    • Mistral Medium 3: 40.0%
    • Mistral Large 4: —

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • Mistral Medium 3: 30.3%
    • Mistral Large 4: —

  • AIME

    Competition mathematics

    • Mistral Medium 3: 44.0%
    • Mistral Large 4: —

  • MATH-500

    Competition math problems

    • Mistral Medium 3: 90.7%
    • Mistral Large 4: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Mistral Medium 3: 3.8%
    • Mistral Large 4: —

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Mistral Medium 3: 24.3%
    • Mistral Large 4: —

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Mistral Medium 3: 31.3%
    • Mistral Large 4: —

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

Live routing data

Performance and reliability

Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.

Throughput
105.1 tok/s
Faster than 64% of 254 models
Time to first token
2.9 s
Quicker than 59% of 254 models
Prompt cache hits
41.1%
Request success (24h)
100.0%
Reliable

Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.

Hourly request success · last 72 hoursLowest hour 100.0%
  • Healthy (98% or more)
  • Degraded (90–98%)
  • Down (under 90%)
  • No requests

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-medium-3",
    "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-medium-3",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Mistral

Other Mistral models, starting with the same family.