Mistral Small vs Nemotron 3 Nano 30B A3B

Mistral Small (Mistral AI) et Nemotron 3 Nano 30B A3B (NVIDIA) côte à côte : prix chez chaque fournisseur, fenêtre de contexte, vitesse et disponibilité mesurées sur Eden AI, fonctionnalités et régions d’hébergement. Les deux s’appellent avec la même clé API Eden AI.

Mistral Small
By
Mistral AI
Input, from (per 1M tokens)
$
0.1
Output, from (per 1M tokens)
$
0.3
Context window
32K
Providers
1
Nemotron 3 Nano 30B A3B
By
NVIDIA
Input, from (per 1M tokens)
$
0.05
Output, from (per 1M tokens)
$
0.2
Context window
262K
Providers
2

Loading the catalog…

Questions about comparing models

Each model is shown at its cheapest provider: per million tokens for language models, and in its own unit for the others (per image, per second of audio, per million characters). Pick a provider under a model to see that provider’s prices instead.
They are measured on real Eden AI traffic. With all of a model’s providers, each figure comes from the provider that did best: last week in the table, the best reading of each day over the last 30 days in the charts. Pick a provider to see its own figures only. A dash means there were too few calls. Only language models are measured.
Some providers of the model keep no data after answering (Zero Data Retention), others may. The figure shows how many of them do. Pick one of those providers to keep every request under Zero Data Retention.
Yes. Every model on this page is called with the same Eden AI API key and the same request format, so switching models means changing one parameter.