Instructions to use leafspark/DeepSeek-V2-Chat-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use leafspark/DeepSeek-V2-Chat-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="leafspark/DeepSeek-V2-Chat-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("leafspark/DeepSeek-V2-Chat-GGUF", dtype="auto")

llama-cpp-python

How to use leafspark/DeepSeek-V2-Chat-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="leafspark/DeepSeek-V2-Chat-GGUF",
	filename="DeepSeek-V2-Chat.bf16.gguf/DeepSeek-V2-Chat.bf16.part-01-of-22.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use leafspark/DeepSeek-V2-Chat-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M

Use Docker

docker model run hf.co/leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use leafspark/DeepSeek-V2-Chat-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "leafspark/DeepSeek-V2-Chat-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "leafspark/DeepSeek-V2-Chat-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M

SGLang

How to use leafspark/DeepSeek-V2-Chat-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "leafspark/DeepSeek-V2-Chat-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "leafspark/DeepSeek-V2-Chat-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "leafspark/DeepSeek-V2-Chat-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "leafspark/DeepSeek-V2-Chat-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use leafspark/DeepSeek-V2-Chat-GGUF with Ollama:
```
ollama run hf.co/leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M
```

Unsloth Studio new

How to use leafspark/DeepSeek-V2-Chat-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for leafspark/DeepSeek-V2-Chat-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for leafspark/DeepSeek-V2-Chat-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for leafspark/DeepSeek-V2-Chat-GGUF to start chatting

Docker Model Runner
How to use leafspark/DeepSeek-V2-Chat-GGUF with Docker Model Runner:
```
docker model run hf.co/leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M
```

Lemonade

How to use leafspark/DeepSeek-V2-Chat-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull leafspark/DeepSeek-V2-Chat-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.DeepSeek-V2-Chat-GGUF-Q4_K_M

List all available models

lemonade list

DeepSeek-V2-Chat-GGUF

Quantizised from https://huggingface.co/deepseek-ai/DeepSeek-V2-Chat

Using llama.cpp b3026 for quantizisation. Given the rapid release of llama.cpp builds, this will likely change over time.

Please set the metadata KV overrides below.

Usage:

Downloading the bf16:

Find the relevant directory
Download all files
Run merge.py
Merged GGUF should appear

Downloading the quantizations:

Find the relevant directory
Download all files
Point to the first split (most programs should load all the splits automatically now)

Running in llama.cpp:

To start in command line chat mode (chat completion):

main -m DeepSeek-V2-Chat.{quant}.gguf -c {context length} --color -c (-i)

To use llama.cpp's OpenAI compatible server:

server \
  -m DeepSeek-V2-Chat.{quant}.gguf \
  -c {context_length} \
  (--color [recommended: colored output in supported terminals]) \
  (-i [note: interactive mode]) \
  (--mlock [note: avoid using swap]) \
  (--verbose) \
  (--log-disable [note: disable logging to file, may be useful for prod]) \
  (--metrics [note: prometheus compatible monitoring endpoint]) \
  (--api-key [string]) \
  (--port [int]) \
  (--flash-attn [note: must be fully offloaded to supported GPU])

Making an importance matrix:

imatrix \
  -m DeepSeek-V2-Chat.{quant}.gguf \
  -f groups_merged.txt \
  --verbosity [0, 1, 2] \
  -ngl {GPU offloading; must build with CUDA} \
  --ofreq {recommended: 1}

Making a quant:

quantize \
  DeepSeek-V2-Chat.bf16.gguf \
  DeepSeek-V2-Chat.{quant}.gguf \
  {quant} \
  (--imatrix [file])

Note: Use iMatrix quants only if you can fully offload to GPU, otherwise speed will be affected negatively.

Quants:

Quant	Status	Size	Description	KV Metadata	Weighted	Notes
BF16	Available	439 GB	Lossless :)	Old	No	Q8_0 is sufficient for most cases
Q8_0	Available	233.27 GB	High quality recommended	Updated	Yes
Q8_0	Available	~110 GB	High quality recommended	Updated	Yes
Q5_K_M	Available	155 GB	Medium-high quality recommended	Updated	Yes
Q4_K_M	Available	132 GB	Medium quality recommended	Old	No
Q3_K_M	Available	104 GB	Medium-low quality	Updated	Yes
IQ3_XS	Available	89.6 GB	Better than Q3_K_M	Old	Yes
Q2_K	Available	80.0 GB	Low quality not recommended	Old	No
IQ2_XXS	Available	61.5 GB	Lower quality not recommended	Old	Yes
IQ1_M	Uploading	27.3 GB	Extremely low quality not recommended	Old	Yes	Testing purposes; use IQ2 at least

Planned Quants (weighted/iMatrix):

Planned Quant	Notes
Q5_K_S
Q4_K_S
Q3_K_S
IQ4_XS
IQ2_XS
IQ2_S
IQ2_M

Metadata KV overrides (pass them using --override-kv, can be specified multiple times):

deepseek2.attention.q_lora_rank=int:1536
deepseek2.attention.kv_lora_rank=int:512
deepseek2.expert_shared_count=int:2
deepseek2.expert_feed_forward_length=int:1536
deepseek2.expert_weights_scale=float:16
deepseek2.leading_dense_block_count=int:1
deepseek2.rope.scaling.yarn_log_multiplier=float:0.0707

License:

DeepSeek license for model weights, which can be found in the LICENSE file in the root of this repo
MIT license for any repo code

Performance:

~1.5t/s with Ryzen 3 3700x (96gb 3200mhz) [Q2_K]

iMatrix:

Find imatrix.dat in the root of this repo, made with a Q2_K quant containing 62 chunks (see here for info: https://github.com/ggerganov/llama.cpp/issues/5153#issuecomment-1913185693)

Using groups_merged.txt, find it here: https://github.com/ggerganov/llama.cpp/discussions/5263#discussioncomment-8395384

Censorship:

This model is a bit censored, finetuning on toxic DPO might help.

Downloads last month: 6,522

GGUF

Model size

236B params

Architecture

deepseek2

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit