apepkuss79's picture
Update README.md with license information (#1)
428647f verified
metadata
base_model: 01-ai/Yi-1.5-34B-Chat
inference: false
model_creator: 01-ai
model_name: Yi-1.5-34B-Chat
model_type: yi
pipeline_tag: text-generation
quantized_by: Second State Inc.
license: apache-2.0

Yi-1.5-34B-Chat-GGUF

Original Model

01-ai/Yi-1.5-34B-Chat

Run with LlamaEdge

  • LlamaEdge version: coming soon

  • Prompt template

    • Prompt type: chatml

    • Prompt string

      <|im_start|>system
      {system_message}<|im_end|>
      <|im_start|>user
      {prompt}<|im_end|>
      <|im_start|>assistant
      
    • Reverse prompt: <|im_end|>

  • Context size: 7168

  • Run as LlamaEdge service

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:Yi-1.5-34B-Chat-Q5_K_M.gguf \
      llama-api-server.wasm \
      --prompt-template chatml \
      --reverse-prompt "<|im_end|>" \
      --ctx-size 7168 \
      --model-name Yi-1.5-34B-Chat
    
  • Run as LlamaEdge command app

    wasmedge --dir .:. --nn-preload default:GGML:AUTO:Yi-1.5-34B-Chat-Q5_K_M.gguf \
      llama-chat.wasm \
      --prompt-template chatml \
      --reverse-prompt "<|im_end|>" \
      --ctx-size 7168
    

Quantized GGUF Models

Name Quant method Bits Size Use case
Yi-1.5-34B-Chat-Q2_K.gguf Q2_K 2 12.8 GB smallest, significant quality loss - not recommended for most purposes
Yi-1.5-34B-Chat-Q3_K_L.gguf Q3_K_L 3 18.1 GB small, substantial quality loss
Yi-1.5-34B-Chat-Q3_K_M.gguf Q3_K_M 3 16.7 GB very small, high quality loss
Yi-1.5-34B-Chat-Q3_K_S.gguf Q3_K_S 3 15 GB very small, high quality loss
Yi-1.5-34B-Chat-Q4_0.gguf Q4_0 4 19.5 GB legacy; small, very high quality loss - prefer using Q3_K_M
Yi-1.5-34B-Chat-Q4_K_M.gguf Q4_K_M 4 20.7 GB medium, balanced quality - recommended
Yi-1.5-34B-Chat-Q4_K_S.gguf Q4_K_S 4 19.6 GB small, greater quality loss
Yi-1.5-34B-Chat-Q5_0.gguf Q5_0 5 23.7 GB legacy; medium, balanced quality - prefer using Q4_K_M
Yi-1.5-34B-Chat-Q5_K_M.gguf Q5_K_M 5 23.4 GB large, very low quality loss - recommended
Yi-1.5-34B-Chat-Q5_K_S.gguf Q5_K_S 5 23.7 GB large, low quality loss - recommended
Yi-1.5-34B-Chat-Q6_K.gguf Q6_K 6 28.3 GB very large, extremely low quality loss
Yi-1.5-34B-Chat-Q8_0.gguf Q8_0 8 36.5 GB very large, extremely low quality loss - not recommended
Yi-1.5-34B-Chat-f16-00001-of-00003.gguf f16 16 32.2 GB
Yi-1.5-34B-Chat-f16-00002-of-00003.gguf f16 16 32.1 GB
Yi-1.5-34B-Chat-f16-00003-of-00003.gguf f16 16 4.48 GB

Quantized with llama.cpp b3135