Mistral 7B [General purpose]

anarkisgaming 's Collections

updated 3 days ago

Mistral 0.2 features 32K context size. It's extremely good at taking the whole context window into account (better than Llama 3 in that regard).

mistralai/Mistral-7B-Instruct-v0.3

Text Generation • Updated 10 days ago • 68.9k • 553

Note Native Context: 32K - Prompt: Mistral - Base: Mistral 0.3 - Function Call Latest version of Mistral 7B
cognitivecomputations/dolphin-2.8-mistral-7b-v02

Text Generation • Updated 12 days ago • 8.71k • 197

Note Native Context: 32K - Prompt: ChatML - Base: Mistral 0.2 raw - Not Aligned Excellent base model for further merge or tuning.
NousResearch/Hermes-2-Pro-Mistral-7B-GGUF

Updated Mar 28 • 25.3k • 211

Note Native Context: 8K (extended to 32k) - Prompt: ChatML - Base: Mistral 0.1 - Function Call Hit and miss due to the subpar version of Mistral being used.