Mixtral 8x7B
Mistral's pioneering MoE open model — one of the original high-quality community models, now superseded.
Pricing
InputOpen weights / ~$0.07 / 1M hosted
OutputOpen weights / ~$0.07 / 1M hosted
Cost bandFree/open
Free tierOpen weights (Apache 2.0).
Limits
Context window32k
Max outputRuntime-dependent
Knowledge cutoffQ3 2023
DeploymentSelf-hosted, Hugging Face, Together, Perplexity, many others
OpenAI-compatible APIYes
ModalitiesText
Where it fits
Best forMature self-hosted stacks that already use Mixtral and haven't migrated to newer open models.
Business fitMigrate to Mistral Small 3.2 or NeMo for better quality and similar hosting cost.
Capabilities
Open weights (Apache 2.0)MoE architectureFunction calling
Benchmark scores
Composite56 / 100
Coding58 / 100 (HumanEval)
Reasoning54 / 100 (MMLU-Pro)
Long context48 / 100
Instruction following60 / 100 (MT-Bench)
Editorial aggregate of published benchmark results as of Mar 1, 2024. These are not independent measurements run by this site.
Change history
No recorded changes for this model yet. Pricing and context are checked daily; changes appear here with dates and sources.
Data
Last checkedJan 10, 2026
Verify with provider documentation before committing spend. Report an error and it gets fixed with a logged correction.