Edit model card

DeepSeek-V2-Lite-Chat-IMat-GGUF

Llama.cpp imatrix quantization of deepseek-ai/DeepSeek-V2-Lite-Chat

Original Model: deepseek-ai/DeepSeek-V2-Lite-Chat
Original dtype: BF16 (bfloat16)
Quantized by: llama.cpp fork PR 7519
IMatrix dataset: here


Files

IMatrix

Status: βœ… Available
Link: here

Common Quants

Filename Quant type File Size Status Uses IMatrix Is Split
DeepSeek-V2-Lite-Chat.Q8_0.gguf Q8_0 16.70GB βœ… Available βšͺ No πŸ“¦ No
DeepSeek-V2-Lite-Chat.Q6_K.gguf Q6_K 14.07GB βœ… Available βšͺ No πŸ“¦ No
DeepSeek-V2-Lite-Chat.Q4_K.gguf Q4_K 10.36GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.Q3_K.gguf Q3_K 8.13GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.Q2_K.gguf Q2_K 6.43GB βœ… Available 🟒 Yes πŸ“¦ No

All Quants

Filename Quant type File Size Status Uses IMatrix Is Split
DeepSeek-V2-Lite-Chat.FP16.gguf F16 31.42GB βœ… Available βšͺ No πŸ“¦ No
DeepSeek-V2-Lite-Chat.BF16.gguf BF16 31.42GB βœ… Available βšͺ No πŸ“¦ No
DeepSeek-V2-Lite-Chat.Q5_K.gguf Q5_K 11.85GB βœ… Available βšͺ No πŸ“¦ No
DeepSeek-V2-Lite-Chat.Q5_K_S.gguf Q5_K_S 11.14GB βœ… Available βšͺ No πŸ“¦ No
DeepSeek-V2-Lite-Chat.Q4_K_S.gguf Q4_K_S 9.53GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.Q3_K_L.gguf Q3_K_L 8.46GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.Q3_K_S.gguf Q3_K_S 7.49GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.Q2_K_S.gguf Q2_K_S 6.46GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ4_NL.gguf IQ4_NL 8.91GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ4_XS.gguf IQ4_XS 8.57GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ3_M.gguf IQ3_M 7.55GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ3_S.gguf IQ3_S 7.49GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ3_XS.gguf IQ3_XS 7.12GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ3_XXS.gguf IQ3_XXS 6.96GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ2_M.gguf IQ2_M 6.33GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ2_S.gguf IQ2_S 6.01GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ2_XS.gguf IQ2_XS 5.97GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ2_XXS.gguf IQ2_XXS 5.64GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ1_M.gguf IQ1_M 5.24GB βœ… Available 🟒 Yes πŸ“¦ No
DeepSeek-V2-Lite-Chat.IQ1_S.gguf IQ1_S 4.99GB βœ… Available 🟒 Yes πŸ“¦ No

Downloading using huggingface-cli

First, make sure you have hugginface-cli installed:

pip install -U "huggingface_hub[cli]"

Then, you can target the specific file you want:

huggingface-cli download legraphista/DeepSeek-V2-Lite-Chat-IMat-GGUF --include "DeepSeek-V2-Lite-Chat.Q8_0.gguf" --local-dir ./

If the model is bigger than 50GB, it will have been split into multiple files. In order to download them all to a local folder, run:

huggingface-cli download legraphista/DeepSeek-V2-Lite-Chat-IMat-GGUF --include "DeepSeek-V2-Lite-Chat.Q8_0/*" --local-dir DeepSeek-V2-Lite-Chat.Q8_0
# see FAQ for merging GGUF's

Inference

Simple chat template

<|begin▁of▁sentence|>User: {user_message_1}

Assistant: {assistant_message_1}<|end▁of▁sentence|>User: {user_message_2}

Assistant:

Chat template with system prompt

<|begin▁of▁sentence|>{system_message}

User: {user_message_1}

Assistant: {assistant_message_1}<|end▁of▁sentence|>User: {user_message_2}

Assistant:

Llama.cpp

llama.cpp/main -m DeepSeek-V2-Lite-Chat.Q8_0.gguf --color -i -p "prompt here (according to the chat template)"

FAQ

Why is the IMatrix not applied everywhere?

According to this investigation, it appears that lower quantizations are the only ones that benefit from the imatrix input (as per hellaswag results).

How do I merge a split GGUF?

  1. Make sure you have gguf-split available
  2. Locate your GGUF chunks folder (ex: DeepSeek-V2-Lite-Chat.Q8_0)
  3. Run gguf-split --merge DeepSeek-V2-Lite-Chat.Q8_0/DeepSeek-V2-Lite-Chat.Q8_0-00001-of-XXXXX.gguf DeepSeek-V2-Lite-Chat.Q8_0.gguf
    • Make sure to point gguf-split to the first chunk of the split.

Got a suggestion? Ping me @legraphista!

Downloads last month
2,800
GGUF
Model size
15.7B params
Architecture
deepseek2

1-bit

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference API (serverless) has been turned off for this model.

Quantized from

Collections including legraphista/DeepSeek-V2-Lite-Chat-IMat-GGUF