Edit model card

Original model: https://huggingface.co/deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct

Tested on Snapdragon X Elite with LM Studio 0.3.2 ARM64 Technology Preview https://lmstudio.ai/snapdragon

Avg answer Speed: 24 tok/s

LM Studio Settings:

Before System: <|im_start|>system\n
After System: <|im_end|>\n
Before User: <|im_start|>user\n
After User: <|im_end|>\n
Before Assistant: <|im_start|>assistant\n
After Assistant: <|im_end|>\n
Downloads last month
138
GGUF
Model size
15.7B params
Architecture
deepseek2

4-bit

Inference Examples
Unable to determine this model's library. Check the docs .

Model tree for pipilok/DeepSeek-Coder-V2-Lite-Instruct-Q4_0_4_8-GGUF

Quantized
(34)
this model