Back to all models
mistralai
mistralaiMistral: Mixtral 8x7B Instruct
Mixtral 8x7B Instruct is a pretrained generative Sparse Mixture of Experts, by Mistral AI, for chat and instruction use. Incorporates 8 experts (feed-forward networks) for a total of 47 billion...
mistralai/mixtral-8x7b-instruct
Context Size
32.768K
Input Price
3,321 Ks/M
Output Price
3,321 Ks/M
Architecture
Text
Supported Parameters
frequency_penaltymax_tokensmin_ppresence_penaltyrepetition_penaltyresponse_formatseedstoptemperaturetool_choicetoolstop_ktop_p
Details
TokenizerMistral
Instruct Typemistral
Max Completion16,384 tokens
Provider Context32.768K tokens
ModeratedNo