Instructions to use karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-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 karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-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 karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16
# Run inference directly in the terminal:
llama cli -hf karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16
# Run inference directly in the terminal:
llama cli -hf karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

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 karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16
# Run inference directly in the terminal:
./llama-cli -hf karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

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 karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

Use Docker

docker model run hf.co/karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

LM Studio
Jan

vLLM

How to use karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-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": "karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

Ollama
How to use karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF with Ollama:
```
ollama run hf.co/karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16
```
Unsloth Desktop

How to use karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

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": "karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF with Docker Model Runner:
```
docker model run hf.co/karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16
```

Lemonade

How to use karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

Run and chat with the model

lemonade run user.Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF-F16

List all available models

lemonade list

Hermes Agent

How to use karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-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 karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

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 karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16

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 "karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF:F16" \
  --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"

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Qwen3.6-35B-A3B Uncensored — karmx mixed-precision GGUF, 128K-tested

GitHub companion repository: KarmSakha/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF — reproducibility files, pipeline scripts, benchmark evidence, and citation information.

Unofficial, calibration-assisted mixed-precision requantization of HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive (itself an abliteration of Qwen/Qwen3.6-35B-A3B), packaged by karmx and validated on a single RTX 5060 Ti 16 GB.

Credit: Qwen team for the base model, HauhauCS for the uncensored variant and source Q8_K_P weights. This is not an official release of either.

Why this exists

The upstream GGUF ladder jumps from IQ2_M (10.86 GiB, ~2.69 bpw) straight to Q2_K_P (13.95 GiB). These two files fill the gap with a tensor-specific mixed recipe: expert down projections (quality-critical) are held at IQ3_XXS, all attention, SSM output and shared-expert tensors at Q4_K, embeddings/output at Q5_K, router at Q6_K, and only the bulk expert gate/up projections drop to the cheapest tier. An imatrix calibrated on a code/tool/multilingual corpus drives the allocation.

Files

File	Bytes	GiB	bpw	Role
`Qwen3.6-35B-A3B-Unc-karmx-mixed-IQ2Q3-128K.gguf`	11,237,388,480	10.47	~2.59	max headroom / longest context
`Qwen3.6-35B-A3B-Unc-karmx-mixed-IQ2MQ3-128K-XL.gguf`	12,579,811,520	11.71	~2.90	quality rung
`mmproj-f16.gguf`	899,283,072	0.84	—	vision projector (unchanged from upstream)

SHA-256 in SHA256SUMS. File size is not VRAM usage; KV cache and compute buffers add on top.

Measured on RTX 5060 Ti 16 GB (Windows, llama.cpp)

Both files served at 131,072 context fully on-GPU (-ngl 99, Q4_0 KV):

Metric	IQ2Q3 (10.47 GiB)	IQ2MQ3-XL (11.71 GiB)
VRAM used @131K	~13.5 GiB	~14.8 GiB
Prefill, 48,314-token prompt	~1,186 tok/s	~1,016 tok/s
Decode @ ~48K depth	~73 tok/s	~73 tok/s
Decode, short context	~93 tok/s	~58–73 tok/s

Hybrid architecture (30 gated-deltanet + 10 full-attention layers, 2 KV heads) keeps the 128K KV cache small and prefill fast. No MTP/nextn block exists in this base, so there is no speculative-decode head to enable.

Run

hf download karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF \
  Qwen3.6-35B-A3B-Unc-karmx-mixed-IQ2MQ3-128K-XL.gguf mmproj-f16.gguf \
  --local-dir ./qwen36-unc

llama-server -m qwen36-unc/Qwen3.6-35B-A3B-Unc-karmx-mixed-IQ2MQ3-128K-XL.gguf \
  --mmproj qwen36-unc/mmproj-f16.gguf --alias qwen36-unc \
  -ngl 99 -c 131072 -np 1 -fa on -b 512 -ub 128 -ctk q4_0 -ctv q4_0 \
  --no-context-shift --jinja --host 127.0.0.1 --port 8080

Pick the IQ2Q3 file if you need the extra ~1.3 GiB headroom (bigger KV type, more slots, or context beyond 131K). Pick XL if you want the extra precision.

Honest limitations

Requantized from upstream Q8_K_P (--allow-requantize), not the BF16 safetensors — a small additional quantization error vs a from-source quant.
imatrix collected on an IQ3_M runner (~3.5 bpw reference), not Q5_K_M.
Calibration corpus (~131K tokens) was built for a sibling 35B-A3B model; same architecture family, but not purpose-built for this exact model.
Tail-chunk expert coverage measured 99.2–99.6% on the last layers, not 100%.
Abliterated/uncensored model — few-to-no refusals by design. Use accordingly.
Perplexity measured on a 12×512-token held-out slice (not a full eval suite); XL showed no measurable PPL gain over base — pick it for headroom-of-mind, not for proven quality.
No guarantee on non-Windows runtimes or other GPUs; numbers are from one card.

Held-out perplexity (llama-perplexity, 12 × 512 tokens, CPU, same corpus slice)

File	bpw	Heldout PPL
karmx IQ2Q3 (10.47 GiB)	2.59	2.2515
karmx IQ2MQ3-XL (11.71 GiB)	2.90	2.2546
upstream IQ3_M (14.38 GiB)	3.56	2.3360

Both karmx rungs land slightly below upstream IQ3_M on this slice despite lower bit-width — consistent with importance-guided allocation spending bits where they matter. Treat single-corpus PPL as a sanity signal, not a full capability evaluation.

Recipe

See CALIBRATION-AND-QUANTIZATION.md and REPRODUCTION.md. Modified-file notice in MODIFICATIONS.txt; attribution in NOTICE.

Downloads last month: 793

GGUF

Model size

35B params

Architecture

qwen35moe

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for karmx/Qwen3.6-35B-A3B-Uncensored-Mixed-128K-GGUF

Base model

Qwen/Qwen3.6-35B-A3B

Quantized

HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive

Quantized

(53)

this model