Difference between revisions of "Mistral AI"
Jump to navigation
Jump to search
| Line 5: | Line 5: | ||
* Products: [[Mistral 7B]], [[Mixtral 8x7B]], [[Mistral Medium]] | * Products: [[Mistral 7B]], [[Mixtral 8x7B]], [[Mistral Medium]] | ||
| + | |||
| + | [[Mistral 7B]] uses [[Grouped-query attention (GQA)]] intended for faster inference and [[Sliding Window Attention (SWA)]] intended to handle longer sequences | ||
| + | |||
| + | == Related == | ||
* <code>[[ollama run mistral]]</code> | * <code>[[ollama run mistral]]</code> | ||
* <code>[[ollama run dolphin-mistral:latest]]</code> | * <code>[[ollama run dolphin-mistral:latest]]</code> | ||
Revision as of 07:10, 25 January 2024
- Products: Mistral 7B, Mixtral 8x7B, Mistral Medium
Mistral 7B uses Grouped-query attention (GQA) intended for faster inference and Sliding Window Attention (SWA) intended to handle longer sequences
Related
See also
Advertising:
