SOLAR-10.7B-Instruct-v1.0-GGUF/README.md

base_model, inference, library_name, license, model_creator, model_name, model_type, datasets, language, pipeline_tag, quantized_by

base_model	inference	library_name	license	model_creator	model_name	model_type	datasets	language	pipeline_tag	quantized_by
upstage/SOLAR-10.7B-Instruct-v1.0	false	transformers	cc-by-nc-4.0	upstage	Solar 10.7B Instruct v1.0	solar	c-s-ale/alpaca-gpt4-data Open-Orca/OpenOrca Intel/orca_dpo_pairs allenai/ultrafeedback_binarized_cleaned	en	text-generation	Second State Inc.

---

SOLAR-10.7B-Instruct-v1.0-GGUF

Original Model

upstage/SOLAR-10.7B-Instruct-v1.0

Run with LlamaEdge

• LlamaEdge version: v0.2.8 and above

• Prompt template

  • Prompt type: solar-instruct

  • Prompt string

    <s> ### User:
    {user_message}
    
    \### Assistant:
    {assistant_message}</s>
    

    Note that the \ character is used to escape the ### in the prompt string. Remove it in the practical use.

• Context size: 4096

• Run as LlamaEdge service

  wasmedge --dir .:. --nn-preload default:GGML:AUTO:SOLAR-10.7B-Instruct-v1.0-Q5_K_M.gguf llama-api-server.wasm -p solar-instruct
  

• Run as LlamaEdge command app

  wasmedge --dir .:. --nn-preload default:GGML:AUTO:SOLAR-10.7B-Instruct-v1.0-Q5_K_M.gguf llama-chat.wasm -p solar-instruct
  
  

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
SOLAR-10.7B-Instruct-v1.0-Q2_K.gguf	Q2_K	2	4.55 GB	smallest, significant quality loss - not recommended for most purposes
SOLAR-10.7B-Instruct-v1.0-Q3_K_L.gguf	Q3_K_L	3	5.65 GB	small, substantial quality loss
SOLAR-10.7B-Instruct-v1.0-Q3_K_M.gguf	Q3_K_M	3	5.19 GB	very small, high quality loss
SOLAR-10.7B-Instruct-v1.0-Q3_K_S.gguf	Q3_K_S	3	4.66 GB	very small, high quality loss
SOLAR-10.7B-Instruct-v1.0-Q4_0.gguf	Q4_0	4	6.07 GB	legacy; small, very high quality loss - prefer using Q3_K_M
SOLAR-10.7B-Instruct-v1.0-Q4_K_M.gguf	Q4_K_M	4	6.46 GB	medium, balanced quality - recommended
SOLAR-10.7B-Instruct-v1.0-Q4_K_S.gguf	Q4_K_S	4	6.1 GB	small, greater quality loss
SOLAR-10.7B-Instruct-v1.0-Q5_0.gguf	Q5_0	5	7.4 GB	legacy; medium, balanced quality - prefer using Q4_K_M
SOLAR-10.7B-Instruct-v1.0-Q5_K_M.gguf	Q5_K_M	5	7.6 GB	large, very low quality loss - recommended
SOLAR-10.7B-Instruct-v1.0-Q5_K_S.gguf	Q5_K_S	5	7.4 GB	large, low quality loss - recommended
SOLAR-10.7B-Instruct-v1.0-Q6_K.gguf	Q6_K	6	8.81 GB	very large, extremely low quality loss
SOLAR-10.7B-Instruct-v1.0-Q8_0.gguf	Q8_0	8	11.4 GB	very large, extremely low quality loss - not recommended