Mistral: 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 parameters. Instruct model fine-tuned by Mistral. #moe
Description
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 parameters. Instruct model fine-tuned by Mistral. #moe
ArchitectureАрхитектура
- Modality:
- text->text
- InputModalities:
- text
- OutputModalities:
- text
- Tokenizer:
- Mistral
- InstructionType:
- mistral
ContextAndLimits
- ContextLength:
- 32768 Tokens
- MaxResponseTokens:
- 16384 Tokens
- Moderation:
- Disabled
PricingRUB
- Request:
- ₽
- Image:
- ₽
- WebSearch:
- ₽
- InternalReasoning:
- ₽
- Prompt1KTokens:
- ₽
- Completion1KTokens:
- ₽
DefaultParameters
- Temperature:
- 0.3
UserComments