wsxiaoys/registry-tabby

🧑‍🔬 Tabby Registry

Completion models (--model)

We recommend using

• For 1B to 3B models, it's advisable to have at least NVIDIA T4, 10 Series, or 20 Series GPUs.
• For 7B to 13B models, we recommend using NVIDIA V100, A100, 30 Series, or 40 Series GPUs.

Model ID	License
wsxiaoys/DeepseekCoder-1.3B	Deepseek License
wsxiaoys/DeepseekCoder-6.7B	Deepseek License
wsxiaoys/DeepseekCoder-v15-7B	Deepseek License
wsxiaoys/StarCoder2-3B	BigCode-OpenRAIL-M
wsxiaoys/StarCoder2-7B	BigCode-OpenRAIL-M
wsxiaoys/StarCoder2-15B	BigCode-OpenRAIL-M
wsxiaoys/CodeGemma-2B	Gemma License
wsxiaoys/CodeGemma-7B	Gemma License
wsxiaoys/CodeQwen-7B	Tongyi Qianwen License
wsxiaoys/DeepSeek-Coder-V2-Lite	Deepseek License
wsxiaoys/StarCoder-1B	BigCode-OpenRAIL-M

Chat models (--chat-model)

To ensure optimal response quality, and given that latency requirements are not stringent in this scenario, we recommend using a model with at least 3B parameters.

Model ID	License
wsxiaoys/Yi-34B	Yi License
wsxiaoys/OpenHermes-2.5-Mistral-7B	Apache 2.0
wsxiaoys/DeepseekCoder-v15-7B-Instruct	Deepseek License
wsxiaoys/DeepseekV2-Lite-Chat	Deepseek License
wsxiaoys/CodeLlama-70B-Instruct	Llama 2
wsxiaoys/CodeGemma-7B-Instruct	Gemma License
wsxiaoys/CodeQwen-7B-Chat	Tongyi Qianwen License
wsxiaoys/Phi-3-mini-128k
wsxiaoys/Qwen2-1.5B-Instruct	Apache 2.0
wsxiaoys/Deepseek-V2-Lite-Chat	Deepseek License
wsxiaoys/Yi-Coder-9B-Chat	Apache 2.0