Instructions to use hermitdave/qwen36-35b-a3b-stq1_0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use hermitdave/qwen36-35b-a3b-stq1_0 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 hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0
# Run inference directly in the terminal:
llama cli -hf hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0
# Run inference directly in the terminal:
llama cli -hf hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0

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 hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0
# Run inference directly in the terminal:
./llama-cli -hf hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0

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 hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0

Use Docker

docker model run hf.co/hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0

LM Studio
Jan
Ollama
How to use hermitdave/qwen36-35b-a3b-stq1_0 with Ollama:
```
ollama run hf.co/hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0
```
Unsloth Desktop

How to use hermitdave/qwen36-35b-a3b-stq1_0 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0

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": "hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use hermitdave/qwen36-35b-a3b-stq1_0 with Docker Model Runner:
```
docker model run hf.co/hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0
```

Lemonade

How to use hermitdave/qwen36-35b-a3b-stq1_0 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0

Run and chat with the model

lemonade run user.qwen36-35b-a3b-stq1_0-TQ1_0

List all available models

lemonade list

Hermes Agent

How to use hermitdave/qwen36-35b-a3b-stq1_0 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0

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 hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use hermitdave/qwen36-35b-a3b-stq1_0 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0

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 "hermitdave/qwen36-35b-a3b-stq1_0:TQ1_0" \
  --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"

Qwen3.6-35B-A3B — STQ1_0 Ternary (Path 1)

A ternary (STQ1_0, 1.31 bpw, 3:4 sparsity) mixed-precision GGUF of Qwen/Qwen3.6-35B-A3B, built by crushing the routed experts on the most salient layers and keeping the parts that shouldn't be crushed at higher precision.

This is the MIX-STQ1_0 recipe from Tencent's Hy4-preview, moved onto Qwen3.6-35B-A3B (35B MoE, 256 experts, hybrid Gated DeltaNet + Gated Attention). Same idea: most of the expert mass goes ternary, the recurrent state and router stay sharp.

Note on the auto-detected tag: Hugging Face reads the filename and labels this TQ1_0. That is misleading. This is a mixed-precision GGUF, not a single TQ1_0 quant. The actual scheme is below.

Quantization map

Tensor group	Quant
Routed experts, layers 0–10 (11 layers)	STQ1_0 (1.31 bpw, ternary)
Routed experts, layers 11–39 (29 layers)	IQ2_XXS
Down-projection experts (all 40 layers)	IQ3_XXS
Gated DeltaNet SSM (`ssm_a / ssm_alpha / ssm_beta / ssm_out`)	Q8_0
MTP head (`nextn.*`)	Q4_K
Router / embeddings / attention	Q8_0 / Q4_K / Q5_K
`output.weight`	bf16 (left as-is)

Result: 2.69 bpw, 11.4 GiB.

What's in the box

qwen36-stq1_0.gguf — 11.4 GiB, 2.69 bpw
qwen36-stq1_0.tensortypes — the per-tensor recipe (read it, it's short)

The recipe, in plain terms

Routed experts: ternary STQ1_0 on the 29 highest-importance layers (picked from bartowski's imatrix), IQ2_XXS on the other 11
Down-projection experts: IQ3_XXS (they write to the residual directly, so they get a bit more)
Gated DeltaNet SSM state (ssm_a / ssm_alpha / ssm_beta / ssm_out): Q8_0, never ternary. The recurrent state is the part you don't want to wreck, and ternary on it quietly destroys long-context recall
MTP head (nextn.*): Q4_K. bartowski's imatrix skips it, so it's pinned by hand
Router, embeddings, attention: Q8_0 / Q4_K / Q5_K as appropriate
output.weight left at bf16

Important: this needs CUDA

STQ1_0's MoE kernels are CUDA-only. The Metal backend in llama.cpp has no STQ1_0 mul_mat_id path, so this file will crash on Apple Silicon. To run it you need a custom llama.cpp build from the pr-22836 branch (the one that adds STQ1_0) with GGML_CUDA=ON. Any CUDA card works — an RTX 3080 with partial offload is fine if you've got the RAM for the rest, though 8 GB VRAM means most layers run on CPU.

llama-cli -m qwen36-stq1_0.gguf -ngl 20 -p "..."

Honest caveats

I could not measure perplexity for this file on the machine I built it on (no CUDA there). The quality story rests on the architecture: the SSM state is protected, which is the main risk. A real quality number needs a CUDA box.
25 GiB is a lot to ship; this repo carries the weight plus the recipe so you can reproduce or adjust the split.

Built as part of a two-path experiment; see the sibling repo qwen36-35b-a3b-p2-iq-mix for an Apple-Silicon-runnable version.

Downloads last month: 300

GGUF

Model size

36B params

Architecture

qwen35moe

Hardware compatibility

1-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for hermitdave/qwen36-35b-a3b-stq1_0

Base model

Qwen/Qwen3.6-35B-A3B

Quantized

(804)

this model