Edit model card
YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

Some GGUF Quants with iMatrix for : https://huggingface.co/NeverSleep/MiquMaid-v2-70B-DPO

Q3_K_M is here, IQ3_XXS on the way.

For other IQ quants (and possibly better ones than mine), see there : https://huggingface.co/Kooten/MiquMaid-v2-70B-DPO-Imatrix-GGUF

Some benchs with LlamaCPP :

  • NeverSleep_MiquMaid-v2-70B-DPO-b2093-iMat-c32_ch1000-Q3_K_M.gguf,-,Hellaswag,83.75,,400,2024-02-07 00:00:00,,70b,Mistral_Medium,32768,,,GGUF,NeverSleep,Nexesenex,

  • NeverSleep_MiquMaid-v2-70B-DPO-b2093-iMat-c32_ch1000-Q3_K_M.gguf,-,Hellaswag,83.2,,1000,2024-02-07 00:00:00,,70b,Mistral_Medium,32768,,,GGUF,NeverSleep,Nexesenex,

  • NeverSleep_MiquMaid-v2-70B-DPO-b2093-iMat-c32_ch1000-Q3_K_M.gguf,-,Arc-Challenge,55.51839465,,299,2024-02-07 05:40:00,,70b,Mistral_Medium,32768,,,GGUF,NeverSleep,Nexesenex,

  • NeverSleep_MiquMaid-v2-70B-DPO-b2093-iMat-c32_ch1000-Q3_K_M.gguf,-,Arc-Easy,78.24561404,,570,2024-02-07 05:40:00,,70b,Mistral_Medium,32768,,,GGUF,NeverSleep,Nexesenex,

  • NeverSleep_MiquMaid-v2-70B-DPO-b2093-iMat-c32_ch1000-Q3_K_M.gguf,-,MMLU,47.92332268,,313,2024-02-07 05:40:00,,70b,Mistral_Medium,32768,,,GGUF,NeverSleep,Nexesenex,

  • NeverSleep_MiquMaid-v2-70B-DPO-b2093-iMat-c32_ch1000-Q3_K_M.gguf,-,Thruthful-QA,40.88127295,,817,2024-02-07 05:40:00,,70b,Mistral_Medium,32768,,,GGUF,NeverSleep,Nexesenex,

  • NeverSleep_MiquMaid-v2-70B-DPO-b2093-iMat-c32_ch1000-Q3_K_M.gguf,-,Winogrande,78.8477,,1267,2024-02-07 05:40:00,,70b,Mistral_Medium,32768,,,GGUF,NeverSleep,Nexesenex,

  • NeverSleep_MiquMaid-v2-70B-DPO-b2093-iMat-c32_ch1000-Q3_K_M.gguf,-,wikitext,4.3328,512,512,2024-02-07 00:00:00,,70b,Mistral_Medium,32768,,,GGUF,NeverSleep,Nexesenex,81

  • NeverSleep_MiquMaid-v2-70B-DPO-b2093-iMat-c32_ch1000-Q3_K_M.gguf,-,wikitext,3.8581,512,512,2024-02-07 00:00:00,70b,Mistral_Medium,32768,,,GGUF,NeverSleep,Nexesenex,655

  • The Hellaswag might be underestimated by 5-6 points due to recent changes in LlamaCPP.

Downloads last month
72
GGUF
Model size
69B params
Architecture
llama

1-bit

3-bit

Inference API
Unable to determine this model's library. Check the docs .