Yi-1.5-9B-Chat-GGUF

Original Model

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: 4096

Run as LlamaEdge service

wasmedge --dir .:. --nn-preload default:GGML:AUTO:Yi-1.5-9B-Chat-Q5_K_M.gguf \
  llama-api-server.wasm \
  --prompt-template chatml \
  --reverse-prompt "<|im_end|>" \
  --ctx-size 4096 \
  --model-name Yi-1.5-9B-Chat

Run as LlamaEdge command app

wasmedge --dir .:. --nn-preload default:GGML:AUTO:Yi-1.5-9B-Chat-Q5_K_M.gguf \
  llama-chat.wasm \
  --prompt-template chatml \
  --reverse-prompt "<|im_end|>" \
  --ctx-size 4096

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
Yi-1.5-9B-Chat-Q2_K.gguf	Q2_K	2	3.35 GB	smallest, significant quality loss - not recommended for most purposes
Yi-1.5-9B-Chat-Q3_K_L.gguf	Q3_K_L	3	4.69 GB	small, substantial quality loss
Yi-1.5-9B-Chat-Q3_K_M.gguf	Q3_K_M	3	4.32 GB	very small, high quality loss
Yi-1.5-9B-Chat-Q3_K_S.gguf	Q3_K_S	3	3.9 GB	very small, high quality loss
Yi-1.5-9B-Chat-Q4_0.gguf	Q4_0	4	5.04 GB	legacy; small, very high quality loss - prefer using Q3_K_M
Yi-1.5-9B-Chat-Q4_K_M.gguf	Q4_K_M	4	5.33 GB	medium, balanced quality - recommended
Yi-1.5-9B-Chat-Q4_K_S.gguf	Q4_K_S	4	5.07 GB	small, greater quality loss
Yi-1.5-9B-Chat-Q5_0.gguf	Q5_0	5	6.11 GB	legacy; medium, balanced quality - prefer using Q4_K_M
Yi-1.5-9B-Chat-Q5_K_M.gguf	Q5_K_M	5	6.26 GB	large, very low quality loss - recommended
Yi-1.5-9B-Chat-Q5_K_S.gguf	Q5_K_S	5	6.11 GB	large, low quality loss - recommended
Yi-1.5-9B-Chat-Q6_K.gguf	Q6_K	6	7.25 GB	very large, extremely low quality loss
Yi-1.5-9B-Chat-Q8_0.gguf	Q8_0	8	9.38 GB	very large, extremely low quality loss - not recommended
Yi-1.5-9B-Chat-f16.gguf	f16	16	17.7 GB

Quantized with llama.cpp b3135