metadata

license: other
pipeline_tag: text-generation
base_model: internlm/internlm2_5-1_8b-chat
model_creator: InternLM
model_name: internlm2_5-1_8b-chat
quantized_by: Second State Inc.

internlm2_5-1_8b-chat-GGUF

Original Model

internlm/internlm2_5-1_8b-chat

Run with LlamaEdge

LlamaEdge version: v0.13.0 and above

Prompt template

Prompt type: chatml

Prompt string

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

Context size: 32000

Run as LlamaEdge service

wasmedge --dir .:. --nn-preload default:GGML:AUTO:internlm2_5-1_8b-chat-Q5_K_M.gguf \
    llama-api-server.wasm \
    --prompt-template chatml \
    --ctx-size 32000 \
    --model-name internlm2_5-1_8b-chat

Run as LlamaEdge command app

wasmedge --dir .:. \
  --nn-preload default:GGML:AUTO:internlm2_5-1_8b-chat-Q5_K_M.gguf \
  llama-chat.wasm \
  --prompt-template chatml \
  --ctx-size 32000

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
internlm2_5-1_8b-chat-Q2_K.gguf	Q2_K	2	722 MB	smallest, significant quality loss - not recommended for most purposes
internlm2_5-1_8b-chat-Q3_K_L.gguf	Q3_K_L	3	1.03 GB	small, substantial quality loss
internlm2_5-1_8b-chat-Q3_K_M.gguf	Q3_K_M	3	964 MB	very small, high quality loss
internlm2_5-1_8b-chat-Q3_K_S.gguf	Q3_K_S	3	888 MB	very small, high quality loss
internlm2_5-1_8b-chat-Q4_0.gguf	Q4_0	4	1.11 GB	legacy; small, very high quality loss - prefer using Q3_K_M
internlm2_5-1_8b-chat-Q4_K_M.gguf	Q4_K_M	4	1.17 GB	medium, balanced quality - recommended
internlm2_5-1_8b-chat-Q4_K_S.gguf	Q4_K_S	4	1.12 GB	small, greater quality loss
internlm2_5-1_8b-chat-Q5_0.gguf	Q5_0	5	1.33 GB	legacy; medium, balanced quality - prefer using Q4_K_M
internlm2_5-1_8b-chat-Q5_K_M.gguf	Q5_K_M	5	1.36 GB	large, very low quality loss - recommended
internlm2_5-1_8b-chat-Q5_K_S.gguf	Q5_K_S	5	1.33 GB	large, low quality loss - recommended
internlm2_5-1_8b-chat-Q6_K.gguf	Q6_K	6	1.55 GB	very large, extremely low quality loss
internlm2_5-1_8b-chat-Q8_0.gguf	Q8_0	8	2.01 GB	very large, extremely low quality loss - not recommended
internlm2_5-1_8b-chat-f16.gguf	f16	16	3.78 GB

Quantized with llama.cpp b3499