M
Mistral Nemo
38
Stack Score
Editorial pick✓ Verified working
Mistral Nemo
Mistral + NVIDIA's 12B open-weights LLM.
Best for
Best for teams who need a capable, license-clean OSS LLM on a single GPU.
At a glance
- Pricing
- open-source
- Setup time
- 30 minutes
- Learning curve
- easy
- Last verified
- 5/26/2026
What we love
- ✓Apache 2.0 — fully permissive
- ✓NVIDIA-optimized inference
- ✓Easy to fit on a single H100
Where it falls short
- −Smaller than top OSS models
- −Behind frontier on hardest tasks
- −Best with NVIDIA stack
Pricing
✓ Pricing verified Jun 21, 20262 tiers
Open-source
$0
perpetual
- ✓Apache 2.0
Hosted
Usage
per token
- ✓Mistral API
- ✓Together
- ✓Fireworks
Overview
Mistral Nemo is a 12B-parameter model jointly built by Mistral and NVIDIA. Strong multilingual support, 128K context, Apache 2.0 license. Punches above its weight class.
Key features
- ●12B parameters
- ●128K context
- ●Apache 2.0 license
- ●Strong multilingual
- ●FP8 inference ready
Best use cases
Single-GPU production
Run a capable LLM on a single GPU without quantization tricks.
Multilingual SaaS
Ship multi-language features that don't embarrass the team.
Integrations
Similar tools
See all LLMs →One AI Stack pick every week.
The tool we’d pay for this week, why we’d pay for it, and what we’d skip. No promo spam.