TinyLlama-1.1B-Chat-v1.0-GGUF

Original Model

Run with LlamaEdge

LlamaEdge version: v0.12.3 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:TinyLlama-1.1B-Chat-v1.0-Q5_K_M.gguf \
  llama-api-server.wasm \
  --prompt-template zephyr \
  --ctx-size 2048 \
  --model-name TinyLlama-1.1B-Chat

Run as LlamaEdge command app

wasmedge --dir .:. --nn-preload default:GGML:AUTO:TinyLlama-1.1B-Chat-v1.0-Q5_K_M.gguf \
  llama-chat.wasm \
  --prompt-template zephyr \
  --ctx-size 2048

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
TinyLlama-1.1B-Chat-v1.0-Q2_K.gguf	Q2_K	2	482 MB	smallest, significant quality loss - not recommended for most purposes
TinyLlama-1.1B-Chat-v1.0-Q3_K_L.gguf	Q3_K_L	3	592 MB	small, substantial quality loss
TinyLlama-1.1B-Chat-v1.0-Q3_K_M.gguf	Q3_K_M	3	550 MB	very small, high quality loss
TinyLlama-1.1B-Chat-v1.0-Q3_K_S.gguf	Q3_K_S	3	499 MB	very small, high quality loss
TinyLlama-1.1B-Chat-v1.0-Q4_0.gguf	Q4_0	4	637 MB	legacy; small, very high quality loss - prefer using Q3_K_M
TinyLlama-1.1B-Chat-v1.0-Q4_K_M.gguf	Q4_K_M	4	668 MB	medium, balanced quality - recommended
TinyLlama-1.1B-Chat-v1.0-Q4_K_S.gguf	Q4_K_S	4	643 MB	small, greater quality loss
TinyLlama-1.1B-Chat-v1.0-Q5_0.gguf	Q5_0	5	766 MB	legacy; medium, balanced quality - prefer using Q4_K_M
TinyLlama-1.1B-Chat-v1.0-Q5_K_M.gguf	Q5_K_M	5	782 MB	large, very low quality loss - recommended
TinyLlama-1.1B-Chat-v1.0-Q5_K_S.gguf	Q5_K_S	5	766 MB	large, low quality loss - recommended
TinyLlama-1.1B-Chat-v1.0-Q6_K.gguf	Q6_K	6	903 MB	very large, extremely low quality loss
TinyLlama-1.1B-Chat-v1.0-Q8_0.gguf	Q8_0	8	1.17 GB	very large, extremely low quality loss - not recommended