metadata

base_model: Qwen/CodeQwen1.5-7B-Chat
inference: false
library_name: transformers
license: other
license_name: tongyi-qianwen
model_creator: Qwen
model_name: Openchat 3.5 0106
model_type: mistral
pipeline_tag: text-generation
quantized_by: Second State Inc.
language:
  - en
tags:
  - chat

CodeQwen1.5-7B-Chat-GGUF

Original Model

Qwen/CodeQwen1.5-7B-Chat

Run with LlamaEdge

LlamaEdge version: v0.9.1 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: 64000

Run as LlamaEdge service

wasmedge --dir .:. --nn-preload default:GGML:AUTO:CodeQwen1.5-7B-Chat-Q5_K_M.gguf \
  llama-api-server.wasm \
  --prompt-template chatml
  --context-size 64000
  --model-name CodeQwen1.5-7B-Chat

Run as LlamaEdge command app

wasmedge --dir .:. --nn-preload default:GGML:AUTO:CodeQwen1.5-7B-Chat-Q5_K_M.gguf \
  llama-chat.wasm \
  --prompt-template chatml \
  --ctx-size 64000

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
CodeQwen1.5-7B-Chat-Q2_K.gguf	Q2_K	2	3.05 GB	smallest, significant quality loss - not recommended for most purposes
CodeQwen1.5-7B-Chat-Q3_K_L.gguf	Q3_K_L	3	3.99 GB	small, substantial quality loss
CodeQwen1.5-7B-Chat-Q3_K_M.gguf	Q3_K_M	3	3.81 GB	very small, high quality loss
CodeQwen1.5-7B-Chat-Q3_K_S.gguf	Q3_K_S	3	3.5 GB	very small, high quality loss
CodeQwen1.5-7B-Chat-Q4_0.gguf	Q4_0	4	4.18 GB	legacy; small, very high quality loss - prefer using Q3_K_M
CodeQwen1.5-7B-Chat-Q4_K_M.gguf	Q4_K_M	4	4.74 GB	medium, balanced quality - recommended
CodeQwen1.5-7B-Chat-Q4_K_S.gguf	Q4_K_S	4	4.41 GB	small, greater quality loss
CodeQwen1.5-7B-Chat-Q5_0.gguf	Q5_0	5	5.04 GB	legacy; medium, balanced quality - prefer using Q4_K_M
CodeQwen1.5-7B-Chat-Q5_K_M.gguf	Q5_K_M	5	5.43 GB	large, very low quality loss - recommended
CodeQwen1.5-7B-Chat-Q5_K_S.gguf	Q5_K_S	5	5.15 GB	large, low quality loss - recommended
CodeQwen1.5-7B-Chat-Q6_K.gguf	Q6_K	6	6.38 GB	very large, extremely low quality loss
CodeQwen1.5-7B-Chat-Q8_0.gguf	Q8_0	8	7.71 GB	very large, extremely low quality loss - not recommended
CodeQwen1.5-7B-Chat-f16.gguf	f16	16	14.5 GB

Quantized with llama.cpp b2781