NOUMENOpen app
NOUMEN

Mistral · Open weights

Devstral 2 123B

Added Dec 9, 2025Open weights

Compare models

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

Loading models…

Use in the app

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.

Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.40
Output
$1.40

Reference prices as published by the model provider. API example

Specifications

Context window
262.1K
Max output
65.5K
Average output (7 days)
120 tokens
Longer than 11% of 468 models

Side by side

Devstral 2 123B 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…

Devstral 2 123BMistralThis modelLFM2.5 2.6BliquidRemove
ProviderMistralliquid
Discovery tierOpen weightsOpen weights
AddedDec 9, 2025Aug 12, 2026
Limits
Context window262.14KBest128K
Max output65.54KBest32.77K
Average output120 tokens459 tokens
Pricing, USD per 1M tokens
Input$0.40$0.10Best
Output$1.40$0.20Best
Performance
Request success (24h)99.7%100.0%Best
Benchmark indexes
Intelligence Index8.6Best8.4
Coding Index31.3Best7.7
Math Index36.7—
Capabilities
ReasoningNoYes
ToolsNoYes
Structured outputNoYes
Reasoning effort—minimal, low, medium, high, xhigh

Public evaluations

Benchmarks

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

  • Devstral 2 123B
  • LFM2.5 2.6B

Intelligence Index

  • Devstral 2 123B8.6
  • LFM2.5 2.6B8.4

Devstral 2 123B: Better than 15% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Devstral 2 123B: 31.3Top
    • LFM2.5 2.6B: 7.7

  • Math Index

    Composite of the math evaluations

    • Devstral 2 123B: 36.7
    • LFM2.5 2.6B: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Devstral 2 123B: 3.1%
    • LFM2.5 2.6B: —

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Devstral 2 123B: 0.0%
    • LFM2.5 2.6B: —

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Devstral 2 123B: 482 Elo
    • LFM2.5 2.6B: —

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Devstral 2 123B: 544 Elo
    • LFM2.5 2.6B: —

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Devstral 2 123B: 2.4%
    • LFM2.5 2.6B: —

  • AA-LCRv1.1

    Reasoning across long documents

    • Devstral 2 123B: 32.3%Top
    • LFM2.5 2.6B: 5.7%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Devstral 2 123B: 3.6%
    • LFM2.5 2.6B: 6.2%Top

  • GPQA Diamond

    Graduate-level science questions

    • Devstral 2 123B: 59.4%Top
    • LFM2.5 2.6B: 55.8%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • Devstral 2 123B: 76.2%
    • LFM2.5 2.6B: —

  • IFBench

    Following precise output instructions

    • Devstral 2 123B: 38.1%
    • LFM2.5 2.6B: —

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Devstral 2 123B: 0.0%
    • LFM2.5 2.6B: —

  • SciCode

    Scientific computing in Python

    • Devstral 2 123B: 32.8%Top
    • LFM2.5 2.6B: 14.4%

  • LiveCodeBench

    Recent competitive programming problems

    • Devstral 2 123B: 44.8%
    • LFM2.5 2.6B: —

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • Devstral 2 123B: 36.7%
    • LFM2.5 2.6B: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Devstral 2 123B: 18.9%
    • LFM2.5 2.6B: —

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Devstral 2 123B: 24.9%
    • LFM2.5 2.6B: —

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Devstral 2 123B: 32.3%Top
    • LFM2.5 2.6B: 5.7%

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.

Prompt cache hits
0.0%
Request success (24h)
99.7%
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 20.0%
  • 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/devstral-2-123b-instruct-2512",
    "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/devstral-2-123b-instruct-2512",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Mistral

Other Mistral models, starting with the same family.