NOUMENOpen app
NOUMEN

Mistral · Open weights

Ministral 14B

Added Dec 4, 2025Open weights

Compare models

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

Loading models…

Use in the app

Ministral 14B is a powerful 14B parameter model from Mistral AI with vision capabilities, offering frontier performance in a compact size.

Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.20
Output
$0.20

Reference prices as published by the model provider. API example

Specifications

Context window
262.1K
Max output
32.8K
Average output (7 days)
176 tokens
Longer than 18% of 468 models

Side by side

Ministral 14B 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…

Ministral 14BMistralThis modelHermes 3 70BnousresearchRemove
ProviderMistralnousresearch
Discovery tierOpen weightsOpen weights
AddedDec 4, 2025Jan 7, 2026
Limits
Context window262.14KBest65.54K
Max output32.77KBest8.19K
Average output176 tokens177 tokens
Pricing, USD per 1M tokens
Input$0.20Best$0.41
Output$0.20Best$0.41
Performance
Throughput—32.4 tok/s
Time to first token—2.0 s
Request success (24h)—100.0%
Benchmark indexes
Intelligence Index6.06.0
Coding Index14.4—
Math Index30.0—
Capabilities
ToolsNoYes
Reasoning effort——

Public evaluations

Benchmarks

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

  • Ministral 14B
  • Hermes 3 70B

Intelligence Index

  • Ministral 14B6.0
  • Hermes 3 70B6.0

Ministral 14B: Better than 3% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Ministral 14B: 14.4
    • Hermes 3 70B: —

  • Math Index

    Composite of the math evaluations

    • Ministral 14B: 30.0
    • Hermes 3 70B: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Ministral 14B: 0.6%
    • Hermes 3 70B: —

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Ministral 14B: 0.2%
    • Hermes 3 70B: —

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Ministral 14B: 131 Elo
    • Hermes 3 70B: —

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Ministral 14B: 247 Elo
    • Hermes 3 70B: —

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Ministral 14B: 1.2%
    • Hermes 3 70B: —

  • AA-LCRv1.1

    Reasoning across long documents

    • Ministral 14B: 26.3%
    • Hermes 3 70B: —

  • MLCR-AA

    Long-context reasoning over medical records

    • Ministral 14B: 2.2%
    • Hermes 3 70B: —

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Ministral 14B: 4.6%Top
    • Hermes 3 70B: 4.0%

  • GPQA Diamond

    Graduate-level science questions

    • Ministral 14B: 57.2%Top
    • Hermes 3 70B: 40.1%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • Ministral 14B: 69.3%Top
    • Hermes 3 70B: 57.1%

  • IFBench

    Following precise output instructions

    • Ministral 14B: 32.0%
    • Hermes 3 70B: —

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Ministral 14B: 0.0%
    • Hermes 3 70B: —

  • SciCode

    Scientific computing in Python

    • Ministral 14B: 23.8%
    • Hermes 3 70B: —

  • LiveCodeBench

    Recent competitive programming problems

    • Ministral 14B: 35.1%Top
    • Hermes 3 70B: 18.8%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • Ministral 14B: 30.0%
    • Hermes 3 70B: —

  • AIME

    Competition mathematics

    • Ministral 14B: —
    • Hermes 3 70B: 2.3%

  • MATH-500

    Competition math problems

    • Ministral 14B: —
    • Hermes 3 70B: 53.8%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Ministral 14B: 4.5%
    • Hermes 3 70B: —

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Ministral 14B: 27.2%
    • Hermes 3 70B: —

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Ministral 14B: 26.3%
    • Hermes 3 70B: —

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/ministral-14b-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/ministral-14b-2512",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Mistral

Other Mistral models, starting with the same family.