Instructions to use huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K
# Run inference directly in the terminal:
llama cli -hf huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K
# Run inference directly in the terminal:
llama cli -hf huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

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 huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K
# Run inference directly in the terminal:
./llama-cli -hf huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

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 huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K
# Run inference directly in the terminal:
./build/bin/llama-cli -hf huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

Use Docker

docker model run hf.co/huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

LM Studio
Jan

vLLM

How to use huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-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": "huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

Ollama
How to use huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF with Ollama:
```
ollama run hf.co/huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K
```
Unsloth Desktop

How to use huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF with Docker Model Runner:
```
docker model run hf.co/huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K
```

Lemonade

How to use huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

Run and chat with the model

lemonade run user.Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF-Q2_K

List all available models

lemonade list

Hermes Agent

How to use huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF:Q2_K" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF / dspark-abliterated /README.md

huihui-ai

Upload folder using huggingface_hub

0698800 verified about 1 month ago

preview code

Raw

History Blame Contribute Delete

4.98 kB

DSpark speculative decoding drafters for DeepSeek-V4-Flash-0731

Preliminary DSpark drafter modules extracted from the official deepseek-ai/DeepSeek-V4-Flash-0731 checkpoint, for use with llama.cpp speculative decoding.

These are optional add-ons. The quants in this repo are unchanged and work exactly as before without them. Nothing in this folder is loaded unless you explicitly ask for it.

Files

File	Size	FP8 source weights	Markov / confidence heads	Routed experts
`dspark-DeepSeek-V4-Flash-0731-BF16.gguf`	11.31 GB	BF16, bit exact	BF16 (source fidelity)	MXFP4 passthrough
`dspark-DeepSeek-V4-Flash-0731-Q8_0.gguf`	10.90 GB	Q8_0	BF16 (source fidelity)	MXFP4 passthrough

Both files contain 81 tensors and are general.architecture = dflash.

Which one: they measured identically in our testing (same acceptance rate, byte identical output text). Take BF16 if you want a provably exact reproduction of DeepSeek's weights, or Q8_0 to save 0.41 GB. There is no measured quality difference between them.

About the names

The 0731 checkpoint stores its drafter as 25 FP8 (E4M3) projections, natively FP4 routed experts, and BF16/F32 for everything small. The two files differ only in how those 25 FP8 tensors are stored:

BF16 upcasts them exactly. FP8 E4M3 carries 4 significant bits and its E8M0 scale is a pure power of two, so BF16 (8 significant bits, full F32 exponent range) reproduces every value with zero error. Verified max|diff| = 0.0 against an independent dequantisation of the source safetensors. This file contains no Q8_0 tensors at all.
Q8_0 stores them as Q8_0, which is what upstream convert_hf_to_gguf.py does by default.

The routed experts are byte identical in both files. They are already MXFP4 in the source checkpoint and are never requantised.

Note that ggml-org/DeepSeek-V4-Flash-0731-GGUF publishes a file also called BF16 which is equivalent to our Q8_0 (its FP8 weights are Q8_0, and only 1.3% of it is actually BF16). Our BF16 is the fully lossless build, which is why it is larger.

Usage

Speculative decoding is opt in. Pass --spec-type draft-dspark or nothing happens.

llama-server \
  -m  huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF/DeepSeek-V4-Flash-Q2-0731.gguf \
  -md huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF/dspark-abliterated/dspark-DeepSeek-V4-Flash-0731-BF16.gguf \
  --spec-type draft-dspark \
  --spec-draft-n-max 5 \
  --fit off \
  -ngl 99 -ngld 99 -fa on -c 8192

Requirements and gotchas:

--fit off is required when using a DSpark drafter.
Do not pass -devd / --spec-draft-device. The drafter ships no token embeddings or output head by design and borrows the target's, so it must span the same devices as the target. Pinning it to one GPU fails with pre-allocated tensor (output.weight) in a buffer (CUDA0) that cannot run the operation.
--spec-draft-n-max is clamped to the trained block size, which is 5 for this model.

Multi GPU requires a rebuild for now

On a device split target, speculative decoding currently aborts on stock builds:

ggml/src/ggml-backend.cpp:1356: GGML_ASSERT(n_graph_inputs < GGML_SCHED_MAX_SPLIT_INPUTS) failed

The DeepSeek-V4 graph needs 39 graph inputs once the speculative layer taps are enabled, and the compile time default is 30. Until this is fixed upstream, rebuild llama.cpp with a higher cap:

cmake -B build -DGGML_CUDA=ON \
  -DCMAKE_CXX_FLAGS="-DGGML_SCHED_MAX_SPLIT_INPUTS=48" \
  -DCMAKE_CUDA_FLAGS="-DGGML_SCHED_MAX_SPLIT_INPUTS=48"

Both flags are needed, otherwise translation units disagree on struct layout. There is no runtime flag or environment variable for this. Single GPU setups are unaffected and work on stock binaries. Raising the cap has no measured throughput or resident memory cost.

This is tracked upstream in the discussion on ggml-org/llama.cpp#25784.

Measured performance

On 4x B200 against UD-Q4_K_XL, greedy, short completions:

	tokens/s	acceptance
target only	62.9	n/a
`--spec-type draft-dspark --spec-draft-n-max 5`	75.6 to 81.9	0.49 to 0.59

That is roughly 1.2x to 1.3x. The gain is workload and hardware dependent; community reports on larger GPU counts and longer generations have seen up to 2x. Acceptance rate is stable across drafter variants but sensitive to prompt and generation length.

Output is not bit identical to non speculative decoding

Speculative decoding should be a pure speed optimisation, but on this model greedy output diverges from a non speculative run of the same prompt. This is a known llama.cpp issue, tracked at ggml-org/llama.cpp#25618, and is not specific to these files.