metadata

base_model: Qwen/Qwen3-0.6B
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-0.6B/blob/main/LICENSE
model_creator: Qwen
model_name: Qwen3-0.6B
quantized_by: Second State Inc.
language:
  - en
pipeline_tag: text-generation
tags:
  - chat

Qwen3-0.6B-GGUF

Original Model

Qwen/Qwen3-0.6B

Run with LlamaEdge

LlamaEdge version:
- Thinking: v0.17.0 and above
- No Thinking: v0.18.2

Prompt template

Prompt type: chatml (for thinking)

Prompt string

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

Prompt type: qwen3-no-think (for no thinking)

Prompt string

<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{user_message_1}<|im_end|>
<|im_start|>assistant
{assistant_message_1}<|im_end|>
<|im_start|>user
{user_message_2}<|im_end|>
<|im_start|>assistant
<think>

</think>

Context size: 128000

Run as LlamaEdge service

wasmedge --dir .:. --nn-preload default:GGML:AUTO:Qwen3-0.6B-Q5_K_M.gguf \
  llama-api-server.wasm \
  --model-name Qwen3-0.6B \
  --prompt-template chatml \
  --ctx-size 128000

Run as LlamaEdge command app

wasmedge --dir .:. --nn-preload default:GGML:AUTO:Qwen3-0.6B-Q5_K_M.gguf \
  llama-chat.wasm \
  --prompt-template chatml \
  --ctx-size 128000

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
Qwen3-0.6B-Q2_K.gguf	Q2_K	2	347 MB	smallest, significant quality loss - not recommended for most purposes
Qwen3-0.6B-Q3_K_L.gguf	Q3_K_L	3	435 MB	small, substantial quality loss
Qwen3-0.6B-Q3_K_M.gguf	Q3_K_M	3	414 MB	very small, high quality loss
Qwen3-0.6B-Q3_K_S.gguf	Q3_K_S	3	390 MB	very small, high quality loss
Qwen3-0.6B-Q4_0.gguf	Q4_0	4	469 MB	legacy; small, very high quality loss - prefer using Q3_K_M
Qwen3-0.6B-Q4_K_M.gguf	Q4_K_M	4	484 MB	medium, balanced quality - recommended
Qwen3-0.6B-Q4_K_S.gguf	Q4_K_S	4	471 MB	small, greater quality loss
Qwen3-0.6B-Q5_0.gguf	Q5_0	5	544 MB	legacy; medium, balanced quality - prefer using Q4_K_M
Qwen3-0.6B-Q5_K_M.gguf	Q5_K_M	5	551 MB	large, very low quality loss - recommended
Qwen3-0.6B-Q5_K_S.gguf	Q5_K_S	5	544 MB	large, low quality loss - recommended
Qwen3-0.6B-Q6_K.gguf	Q6_K	6	623 MB	very large, extremely low quality loss
Qwen3-0.6B-Q8_0.gguf	Q8_0	8	805 MB	very large, extremely low quality loss - not recommended
Qwen3-0.6B-f16.gguf	f16	16	1.51 GB

Quantized with llama.cpp b5097