ZMIME
Mistral AI · Released Jul 18, 2024

Mistral Nemo

Efficient Mistral-NVIDIA open model for multilingual chat and local deployment.

  • Open weights
  • Tool calling
Capability
113.8ECI · #155 of 158
Input
$0.09per 1M tokens
Output
$0.109per 1M tokens
Context
128K128K max output
Pricing

What it costs

Input
$0.09per million tokens
Output
$0.109per million tokens
Blended (3:1)
$0.095cheaper than 91% of priced models

Typical monthly bills

Monthly usageEstimated cost
Side project2M input + 0.5M output tokens$0.235
Team assistant25M input + 5M output tokens$2.80
Production app250M input + 50M output tokens$28.01

Median of 11 paid API providers on models.dev.

Capability

Independent benchmarks

113.8Epoch Capabilities Index
#155 of 158 scored models
68Median 147174

The shaded band is Epoch AI’s confidence range (108.8–118.2); the tick marks the median scored model.

  • GPQA DiamondGraduate-level science questions29.9%

Scores from Epoch AI, run independently of Mistral AI.

Specifications

The details

Lab
Mistral AI
Released
Jul 18, 2024
Knowledge cutoff
Jul 2024
Context window
128,000 tokens
Max output
128,000 tokens
Inputs
Text
Output
Text
Reasoning
No
Tool calling
Yes
Structured output
No
Availability
11 API providerslisted on models.dev
Same lab

More from Mistral AI

Questions

About Mistral Nemo

How much does Mistral Nemo cost?

Mistral Nemo costs $0.09 input / $0.109 output per million tokens (median across 11 API providers). At a 3:1 input-to-output mix that is $0.095 per million tokens, cheaper than 91% of the 469 priced models we track.

What is the context window of Mistral Nemo?

Mistral Nemo accepts up to 128,000 tokens per request and can write up to 128,000 tokens in one response.

How good is Mistral Nemo?

Epoch AI gives Mistral Nemo a Capabilities Index score of 113.8 (likely range 108.8–118.2), ranking it #155 of 158 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 29.9%.

Is Mistral Nemo open source?

Yes. Mistral AI publishes the weights, so it can be downloaded and self-hosted.

What inputs does Mistral Nemo support?

Mistral Nemo accepts text and replies in text. It is not a dedicated reasoning model, supports tool calling.

When was Mistral Nemo released?

Mistral AI released Mistral Nemo on Jul 18, 2024. Its training data runs to Jul 2024.

What are the best alternatives to Mistral Nemo?

The closest current models from other labs on capability, price and release date are Llama-3.1-8B-Instruct (Meta, ECI 110.8, $0.055 / $0.078), Qwen2.5 7B Instruct (Alibaba (Qwen), $0.075 / $0.15), Claude Haiku 3 (Anthropic, ECI 117.8, $0.25 / $1.25) and Gemma 2 27B (Google, ECI 118.6, $0.65 / $0.65).