metadata

base_model: HuggingFaceH4/zephyr-7b-beta
inference: false
license: mit
model_creator: Hugging Face H4
model_name: Zephyr 7B Beta
model_type: mistral
pipeline_tag: text-generation
quantized_by: Second State Inc.
tags:
  - generated_from_trainer
datasets:
  - HuggingFaceH4/ultrachat_200k
  - HuggingFaceH4/ultrafeedback_binarized
language:
  - en

Zephyr-7B-Beta-GGUF

Original Model

HuggingFaceH4/zephyr-7b-beta

Run with LlamaEdge

LlamaEdge version: v0.2.8 and above

Prompt template

Prompt type: zephyr

Prompt string

<|system|>
{system_prompt}</s>
<|user|>
{prompt}</s>
<|assistant|>

Reverse prompt: </s>

Context size: 4096

Run as LlamaEdge service

wasmedge --dir .:. --nn-preload default:GGML:AUTO:zephyr-7b-beta-Q5_K_M.gguf llama-api-server.wasm -p zephyr -r '</s>'

Run as LlamaEdge command app

wasmedge --dir .:. --nn-preload default:GGML:AUTO:zephyr-7b-beta-Q5_K_M.gguf llama-chat.wasm -p zephyr -r '</s>'

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
zephyr-7b-beta-Q2_K.gguf	Q2_K	2	3.08 GB	smallest, significant quality loss - not recommended for most purposes
zephyr-7b-beta-Q3_K_L.gguf	Q3_K_L	3	3.82 GB	small, substantial quality loss
zephyr-7b-beta-Q3_K_M.gguf	Q3_K_M	3	3.52 GB	very small, high quality loss
zephyr-7b-beta-Q3_K_S.gguf	Q3_K_S	3	3.16 GB	very small, high quality loss
zephyr-7b-beta-Q4_0.gguf	Q4_0	4	4.11 GB	legacy; small, very high quality loss - prefer using Q3_K_M
zephyr-7b-beta-Q4_K_M.gguf	Q4_K_M	4	4.37 GB	medium, balanced quality - recommended
zephyr-7b-beta-Q4_K_S.gguf	Q4_K_S	4	4.14 GB	small, greater quality loss
zephyr-7b-beta-Q5_0.gguf	Q5_0	5	5.00 GB	legacy; medium, balanced quality - prefer using Q4_K_M
zephyr-7b-beta-Q5_K_M.gguf	Q5_K_M	5	5.13 GB	large, very low quality loss - recommended
zephyr-7b-beta-Q5_K_S.gguf	Q5_K_S	5	5.00 GB	large, low quality loss - recommended
zephyr-7b-beta-Q6_K.gguf	Q6_K	6	5.94 GB	very large, extremely low quality loss
zephyr-7b-beta-Q8_0.gguf	Q8_0	8	7.70 GB	very large, extremely low quality loss - not recommended