NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Mistral Nemo
Mistral AI logo

Mistral Nemo

by Mistral AI · Released Jul 2024

Open Source
35.4
avg score
Rank #223
Compare
Better than 29% of all models
Context
131K tokens (~66 books)
Input $/1M
$0.02
Output $/1M
$0.03
Type
text
License
Open Source
Benchmarks
6 tested
Data as of
About

A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA. The model is multilingual, supporting English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese,...

Tested on 6 benchmarks · BenchGecko score 35.4. Top scores: GSM8K (84.2%), PIQA (67.0%), Balrog (17.6%).

Looking for similar performance at lower cost?
Gemma 3 27B (free) scores 35.1 (99% as good) at $0.00/1M input · 100% cheaper
Capabilities
math
47.5
#124 globally
knowledge
30.4
#241 globally
general
14.3
#181 globally
Benchmark Scores
Compare All
Tested on 6 benchmarks · Ranked across 3 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
mathCompare math →
GSM8K

Grade school math word problems. 8,500 problems testing multi-step arithmetic reasoning. A foundational math benchmark.

84.2·
MATH level 5

Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.

10.8·
knowledgeCompare knowledge →
PIQA

Physical Intuition QA. Tests understanding of everyday physical interactions and commonsense physics.

67.0·
Balrog

Broad Assessment of Language and Reasoning Over Games. Tests strategic and logical reasoning through game scenarios.

17.6·
GPQA diamond

Graduate-level science questions written by PhD experts. Diamond subset contains questions where experts disagree, testing deep understanding.

6.5·
generalCompare general →
Dtbench
14.3·
Quick compare:Meta logovs Llama 3.2 3B InstructOpenAI logovs GPT-4.1 MiniMistral AI logovs Mistral Large 2407
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Recently Happened
Mistral Nemo pricing dropped 25%
Apr 29, 2026
Mistral Nemo pricing dropped 50%
Apr 24, 2026
Similar Models
Meta logo
Llama 3.2 3B Instruct
Meta
35.4$0.05/1M
OpenAI logo
GPT-4.1 Mini
OpenAI
35.2$0.40/1M
Mistral AI logo
Mistral Large 2407
Mistral AI
35.1$2.00/1M
Links
Info
Mistral AIPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Mistral AI
Source Code
GitHubHugging Face
BenchGecko API
mistral-nemo
Specifications
  • Typetext
  • Context131K tokens (~66 books)
  • ReleasedJul 2024
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.000
Available On
Mistral AI logoMistral AI$0.02
Categories
mathknowledgegeneral
Learn More
context-windowtransformeropen weightstokens
Share & Export
Tweet
Related Models
Meta logoLlama 3.2 3B Instruct
35.4
OpenAI logoGPT-4.1 Mini
35.2
Mistral AI logoMistral Large 2407
35.1
Google DeepMind logoGemma 3 27B (free)
35.1
Mistral Nemo is an open-source text AI model by Mistral AI, released in July 2024. It has an average benchmark score of 35.4. Context window: 131K tokens.

Related Models

Llama 3.2 3B Instruct · MetaGPT-4.1 Mini · OpenAIMistral Large 2407 · Mistral AIGemma 3 27B (free) · Google DeepMindMistral Large 2411 · Mistral AI

Benchmarks

GSM8KPIQABalrogDtbenchMATH level 5

Related Pages

Mistral AI · ProviderMistral AI · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-10-05

  • Mistral Nemo by Mistral AI. BenchGecko score 35.4, rank 223 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.0190 input · $0.0300 output per 1M tokens (as of 2026-10-05).
  • Sold by 6 providers (as of 2026-10-05): DekaLLM (fp8) $0.0180 in / $0.0300 out · DeepInfra (fp8) $0.0190 in / $0.0300 out · Parasail (fp8) $0.0300 in / $0.0300 out · Io Net (fp16) $0.0321 in / $0.12 out · Novita (fp8) $0.0400 in / $0.17 out · and 1 more. Every provider

How to cite · data as of 2026-10-05

Mistral Nemo · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/mistral-nemo

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP