Instructions to use 6block/DeepSeek-V4-Pro-0813-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 6block/DeepSeek-V4-Pro-0813-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 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S
# Run inference directly in the terminal:
llama cli -hf 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S
# Run inference directly in the terminal:
llama cli -hf 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

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 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S
# Run inference directly in the terminal:
./llama-cli -hf 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

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 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S
# Run inference directly in the terminal:
./build/bin/llama-cli -hf 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

Use Docker

docker model run hf.co/6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

LM Studio
Jan

vLLM

How to use 6block/DeepSeek-V4-Pro-0813-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "6block/DeepSeek-V4-Pro-0813-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": "6block/DeepSeek-V4-Pro-0813-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

Ollama
How to use 6block/DeepSeek-V4-Pro-0813-GGUF with Ollama:
```
ollama run hf.co/6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S
```

Unsloth Studio

How to use 6block/DeepSeek-V4-Pro-0813-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 6block/DeepSeek-V4-Pro-0813-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 6block/DeepSeek-V4-Pro-0813-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for 6block/DeepSeek-V4-Pro-0813-GGUF to start chatting

How to use 6block/DeepSeek-V4-Pro-0813-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use 6block/DeepSeek-V4-Pro-0813-GGUF with Docker Model Runner:
```
docker model run hf.co/6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S
```

Lemonade

How to use 6block/DeepSeek-V4-Pro-0813-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

Run and chat with the model

lemonade run user.DeepSeek-V4-Pro-0813-GGUF-IQ1_S

List all available models

lemonade list

Hermes Agent

How to use 6block/DeepSeek-V4-Pro-0813-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 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

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 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use 6block/DeepSeek-V4-Pro-0813-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf 6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S

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 "6block/DeepSeek-V4-Pro-0813-GGUF:IQ1_S" \
  --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"

DeepSeek-V4-Pro-0813 GGUF

Sub-4-bit GGUF quantizations of deepseek-ai/DeepSeek-V4-Pro-0813, produced by the 6block team with importance-matrix (imatrix) calibration.

1.57T parameters, 48B active per token. 61 layers, 384 routed experts (top-6) + 1 shared expert.

Why only sub-4-bit tiers

The upstream weights ship in FP4 (expert_dtype: fp4 in config.json). Routed-expert tensors are stored pre-packed, so convert_hf_to_gguf.py writes them straight to GGUF's MXFP4 type without ever materialising BF16. The resulting F16 GGUF is 812.7 GiB at 4.33 bpw, and expert tensors are 96.4% of it.

That means the usual "higher tier = better" ladder does not apply. Measured with llama-quantize --dry-run against this exact master:

Tier	Size	vs master	Verdict
Q8_0	1556.7 GiB	+96%	inflates, no quality gained
Q6_K	1201.9 GiB	+48%	inflates
Q5_K_M	1038.9 GiB	+28%	inflates
Q4_K_M	885.6 GiB	+12%	inflates
IQ4_XS	787.4 GiB	−3%	not worth publishing

Quantizing a 4.25-bpw tensor up to 8 bpw only doubles the file; it cannot recover information the factory FP4 step already discarded. Everything published here is below the master's 4.33 bpw.

For 4-bit and 8-bit builds of this model, see unsloth/DeepSeek-V4-Pro-0813-GGUF (UD-Q4_K_XL 850 GB, UD-Q8_K_XL 873 GB). This repository covers the range below that.

Available quantizations

Tier	Size	bpw	PPL (12 chunks)	Notes
Q3_K_M	711.3 GiB	3.88	1.6217 ± 0.0528	highest quality here
IQ3_XXS	577.0 GiB	3.15	1.6708 ± 0.0547	best size/quality balance
Q2_K	547.0 GiB	2.99	1.7621 ± 0.0594
IQ1_M	346.6 GiB	1.89	3.6966 ± 0.1640	quality drops sharply
IQ1_S	314.1 GiB	1.72	4.1095 ± 0.1799	smallest

Master F16 GGUF baseline: PPL 4.0795 ± 0.0458 (measured during imatrix, 220 chunks — a different chunk count than the table above, so it is not directly comparable; see caveats).

The quality cliff sits between Q2_K and IQ1_M: 200 GiB of savings costs +1.93 PPL, whereas the entire Q3_K_M → Q2_K range costs only +0.14.

Tiers deliberately not published

IQ2_XS (445 GiB, PPL 4.4705) and IQ2_XXS (401 GiB, PPL 21.6798) were built and then rejected. Both are beaten outright by smaller files — IQ1_S is 315 GiB at PPL 4.11 — so they occupy a size bracket while delivering worse output. The IQ2 expert-quantization path appears to break down on this sparse-routing MoE; the same failure mode showed up on DeepSeek-V4-Flash's IQ2_M. Sizes and tensor counts looked completely normal, which is why every tier here was PPL-tested before release.

Quantization details

Tool: llama.cpp @ 4ed2b13 (needs LLM_ARCH_DEEPSEEK4; older builds reject deepseek4)
imatrix: 220 chunks over a 476 KB multilingual corpus (EN/ZH), final PPL 4.0795, published as imatrix.gguf
Requantization: --allow-requantize is mandatory. Expert tensors arrive already quantized as MXFP4, and llama.cpp refuses to requantize by default (requantizing from type mxfp4 is disabled). Note this makes every tier here a second quantization pass on top of the factory FP4 step.
Non-expert tensors are protected explicitly, because a global low-bit setting would otherwise crush the sparse-attention indexer and the per-layer control tensors:
```
hc_*                    → F32     (per-layer control)
attn_q/k/v/output       → Q8_0
indexer*, compressor*   → Q8_0    (sparse-attention index path)
ffn_gate_inp            → F32     (router)
shexp                   → Q8_0    (shared expert)
token_embd, output      → Q6_K
```
--tensor-type matches substrings and first match wins, so attn_ alone would also swallow hc_attn_fn. The four attention projections are listed separately on purpose.
Metadata: general.quantized_by=6block, no absolute paths in any KV field.

Usage

llama.cpp

hf download 6block/DeepSeek-V4-Pro-0813-GGUF \
  DeepSeek-V4-Pro-0813-IQ3_XXS.gguf --local-dir .

llama-server -m DeepSeek-V4-Pro-0813-IQ3_XXS.gguf -c 8192 --jinja

Do not pass -ngl or --n-cpu-moe manually. Setting either makes llama.cpp abandon automatic VRAM fitting and split by layer count instead, which overflows individual cards on a model this size (common_fit_params: n_gpu_layers already set by user to 99, abort, then cudaMalloc failed). Let it fit the model itself.

Ollama

cat > Modelfile <<'EOF'
FROM ./DeepSeek-V4-Pro-0813-IQ3_XXS.gguf
PARAMETER temperature 0.6
PARAMETER top_p 0.95
EOF

ollama create deepseek-v4-pro -f Modelfile
ollama run deepseek-v4-pro

Caveats

Read these before comparing numbers with any other repository.

PPL is wikitext-2, n_ctx=512, 12 chunks. Cross-tier comparisons in the table are valid; comparisons against other models or other repos' published figures are not. Perplexity's running average climbs monotonically as more corpus is covered, so a 12-chunk number and a 568-chunk number are different measurements even for the same file.
The 4.0795 master baseline was measured at 220 chunks, during the imatrix pass — not at 12. It indicates the master's general range, not a like-for-like delta against the table.
Every tier is a double quantization (factory FP4 → MXFP4 → target). Losses appear smaller than they would from a BF16 master, because the first pass already removed most of the information. That is a property of this master, not evidence of a better recipe.
PPL is not generation quality. It measures language-modelling loss on one English corpus. The 1-bit tiers pass the numeric gate but have not been evaluated for instruction following, long-context behaviour, or agentic use. Test before deploying.
No benchmark suite was run. No MMLU, GSM8K, or coding evaluations — only perplexity.

License

MIT, inherited from the upstream model. See the original repository for terms.

Quantized by the 6block team.

Downloads last month: -

GGUF

Model size

1.6T params

Architecture

deepseek4

Hardware compatibility

1-bit

View +1 variant

Model tree for 6block/DeepSeek-V4-Pro-0813-GGUF

Base model

deepseek-ai/DeepSeek-V4-Pro-0813

Quantized

(8)

this model