second-state/SmolLM-135M-Instruct-GGUF

---

SmolLM-135M-Instruct

Original Model

HuggingFaceTB/SmolLM-135M-Instruct

Run with LlamaEdge

• LlamaEdge version: v0.12.5 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: 2048

• Run as LlamaEdge service

  wasmedge --dir .:. --nn-preload default:GGML:AUTO:SmolLM-135M-Instruct-Q5_K_M.gguf \
    llama-api-server.wasm \
    --prompt-template chatml \
    --ctx-size 2048 \
    --model-name SmolLM-135M-Instruct
  

• Run as LlamaEdge command app

  wasmedge --dir .:. --nn-preload default:GGML:AUTO:SmolLM-135M-Instruct-Q5_K_M.gguf \
    llama-chat.wasm \
    --prompt-template chatml \
    --ctx-size 2048
  

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
SmolLM-135M-Instruct-Q2_K.gguf	Q2_K	2	88.2 MB	smallest, significant quality loss - not recommended for most purposes
SmolLM-135M-Instruct-Q3_K_L.gguf	Q3_K_L	3	97.5 MB	small, substantial quality loss
SmolLM-135M-Instruct-Q3_K_M.gguf	Q3_K_M	3	93.5 MB	very small, high quality loss
SmolLM-135M-Instruct-Q3_K_S.gguf	Q3_K_S	3	88.2 MB	very small, high quality loss
SmolLM-135M-Instruct-Q4_0.gguf	Q4_0	4	91.7 MB	legacy; small, very high quality loss - prefer using Q3_K_M
SmolLM-135M-Instruct-Q4_K_M.gguf	Q4_K_M	4	105 MB	medium, balanced quality - recommended
SmolLM-135M-Instruct-Q4_K_S.gguf	Q4_K_S	4	102 MB	small, greater quality loss
SmolLM-135M-Instruct-Q5_0.gguf	Q5_0	5	105 MB	legacy; medium, balanced quality - prefer using Q4_K_M
SmolLM-135M-Instruct-Q5_K_M.gguf	Q5_K_M	5	112 MB	large, very low quality loss - recommended
SmolLM-135M-Instruct-Q5_K_S.gguf	Q5_K_S	5	110 MB	large, low quality loss - recommended
SmolLM-135M-Instruct-Q6_K.gguf	Q6_K	6	138 MB	very large, extremely low quality loss
SmolLM-135M-Instruct-Q8_0.gguf	Q8_0	8	145 MB	very large, extremely low quality loss - not recommended
SmolLM-135M-Instruct-f16.gguf	f16	16	271 MB

Quantized with llama.cpp b3445.