Pixtral 12B
Compact Vision
Smaller open-weights Pixtral. Single-GPU vision model from Mistral; the open-weights answer to GPT-4o-mini class multimodal.
Pixtral 12B is a multimodal AI model from Mistral. It costs $0.150 per million input tokens and $0.150 per million output tokens (blended $0.150/M), with a 128,000-token context window.
INPUT
$0.150/M
per million input tokens
OUTPUT
$0.150/M
per million output tokens
CONTEXT
128,000
tokens
What it is good at
- Open-weights vision
- Single-GPU
- Apache 2.0 license
Typical use cases
- Self-hosted vision QA
- On-prem multimodal
- Fine-tune base
Benchmarks
vs. best public score
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources, treat as directional rather than authoritative.
More from Mistral
See all 15 →Mistral Large 3
Frontier · 256,000 ctx
in $0.500/Mout $1.500/M
Mistral Medium 3.5
Balanced · 256,000 ctx
in $1.500/Mout $7.500/M
Mistral Small 4
Efficient · 128,000 ctx
in $0.100/Mout $0.300/M
Mistral Large 2
Frontier · 128,000 ctx
in $2.000/Mout $6.000/M
Pixtral Large
Vision · 128,000 ctx
in $2.000/Mout $6.000/M
Mistral Small 3.2
Efficient · 128,000 ctx
in $0.100/Mout $0.300/M
Frequently asked questions
How much does Pixtral 12B cost?
Pixtral 12B costs $0.150 per million input tokens and $0.150 per million output tokens, for a blended reference rate of $0.150 per million tokens.
What is Pixtral 12B's context window?
Pixtral 12B supports up to 128,000 tokens of context in a single request.
What is Pixtral 12B best for?
Pixtral 12B is well suited to Open-weights vision, Single-GPU and Apache 2.0 license.
Who makes Pixtral 12B?
Pixtral 12B is developed and served by Mistral. It was released in Sep 2024.