Supported Models

Chat LLM

Name	HuggingFace Identifier	배포 추가 설정	supported roles	vRAM Size
Gemma Instruction	casperhansen/gemma-7b-it-awq	quantization=awq	assitant, user
LM Sys	TheBloke/vicuna-13B-v1.5-AWQ	quantization=awq	system, assitant, user
LLaMA-2 Chat	TheBloke/Llama-2-13B-chat-AWQ	quantization=awq	system, assitant, user
LLaMA-3 Chat	Sreenington/Llama-3-8B-ChatQA-AWQ	quantization=awq	system, assitant, user
LLaMA-2 Instruct	TheBloke/Llama-2-7B-32K-Instruct-AWQ	quantization=awq	system, assitant, user
LLaMA-3 Instruct	solidrust/Meta-Llama-3-8B-Instruct-AWQ	quantization=awq	system, assitant, user
Code Llama Instruct	solidrust/Llama-3-8B-Instruct-Coder-AWQ	quantization=awq	system, assitant, user
Mistral (7B) Instruct	TheBloke/Mistral-7B-Instruct-v0.1-AWQ	quantization=awq, dtype=half	system, assitant, user
Mistral (7B) Instruct v0.2	TheBloke/Mistral-7B-Instruct-v0.2-AWQ	quantization=awq	assitant, user
Mistral (7B) Instruct v0.3	solidrust/Mistral-7B-Instruct-v0.3-AWQ	quantization=awq	assitant, user
OpenChat 3.5	TheBloke/openchat_3.5-AWQ	quantization=awq	system, assitant, user
Solar Instruct	TheBloke/SOLAR-10.7B-Instruct-v1.0-AWQ	quantization=awq	system, assitant, user
EEVE Instruct	yanolja/EEVE-Korean-Instruct-10.8B-v1.0	quantization=awq, dtype=half	system, assitant, user
Qwen2 Instruct	Qwen/Qwen2-7B-Instruct-AWQ	quantization=awq	system, assitant, user
Qwen 1.5 Instruct	Qwen/Qwen2-1.5B-Instruct-AWQ	quantization=awq	system, assitant, user
Phi-3 Instruct	microsoft/Phi-3-mini-4k-instruct		system, assitant, user
Yi Chat(9B)	modelscope/Yi-1.5-9B-Chat-AWQ	quantization=awq	system, assitant, user

Completion LLM

Name	Supported	HuggingFace Identifier	배포 추가 설정	vRAM Size
Gemma	O	TechxGenus/gemma-7b-AWQ	quantization=awq
LLaMA-2	O	TheBloke/Llama-2-7B-AWQ	quantization=awq
LLaMA-3	O	solidrust/Meta-Llama-3-8B-AWQ	quantization=awq
Qwen2	X	macadeliccc/Samantha-Qwen2-7B-AWQ	quantization=awq
Mistral	X	TheBloke/Mistral-7B-v0.1-AWQ	quantization=awq
01-ai Yi Base (6B)	O	Ichigo2899/01-ai-Yi-1.5-9B-AWQ	quantization=awq
SOLAR-10.7B	O	TheBloke/SOLAR-10.7B-v1.0-AWQ	quantization=awq
EEVE	O	yanolja/EEVE-Korean-10.8B-v1.0	max-model-len=2576

Image Models

Model Name	Supported	Text-to-Image	Image-to-Image	HuggingFace Identifier	vRAM Size
Stable Diffusion 1.5	O	O	X	runwayml/stable-diffusion-v1-5
Stable Diffusion 2	O	O	X	stabilityai/stable-diffusion-2
Stable Diffusion 2.1	O	O	X	stabilityai/stable-diffusion-2-1
Stable Diffusion XL	O	O	O	vae model: madebyollin/sdxl-vae-fp16-fix base model: stabilityai/stable-diffusion-xl-base-1.0 refiner model: stabilityai/stable-diffusion-xl-refiner-1.0
Analog Diffusion	X	O	X	wavymulder/Analog-Diffusion
Openjourney	O	O	X	prompthero/openjourney
Realistic Vision	O	O	X	SG161222/Realistic_Vision_V3.0_VAE