Instructions to use nerozx/qwen3.5-2b-decision-browser-q4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use nerozx/qwen3.5-2b-decision-browser-q4 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 nerozx/qwen3.5-2b-decision-browser-q4:Q8_0 # Run inference directly in the terminal: llama cli -hf nerozx/qwen3.5-2b-decision-browser-q4:Q8_0
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf nerozx/qwen3.5-2b-decision-browser-q4:Q8_0 # Run inference directly in the terminal: llama cli -hf nerozx/qwen3.5-2b-decision-browser-q4:Q8_0
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf nerozx/qwen3.5-2b-decision-browser-q4:Q8_0 # Run inference directly in the terminal: ./build/bin/llama cli -hf nerozx/qwen3.5-2b-decision-browser-q4:Q8_0
Use Docker
docker model run hf.co/nerozx/qwen3.5-2b-decision-browser-q4:Q8_0
- Llama new
- LM Studio
- Jan
- vLLM
How to use nerozx/qwen3.5-2b-decision-browser-q4 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "nerozx/qwen3.5-2b-decision-browser-q4" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "nerozx/qwen3.5-2b-decision-browser-q4", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/nerozx/qwen3.5-2b-decision-browser-q4:Q8_0
- Ollama
How to use nerozx/qwen3.5-2b-decision-browser-q4 with Ollama:
ollama run hf.co/nerozx/qwen3.5-2b-decision-browser-q4:Q8_0
- Unsloth Desktop
- Pi
How to use nerozx/qwen3.5-2b-decision-browser-q4 with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf nerozx/qwen3.5-2b-decision-browser-q4:Q8_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:9931/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "nerozx/qwen3.5-2b-decision-browser-q4:Q8_0" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use nerozx/qwen3.5-2b-decision-browser-q4 with Docker Model Runner:
docker model run hf.co/nerozx/qwen3.5-2b-decision-browser-q4:Q8_0
- Lemonade
How to use nerozx/qwen3.5-2b-decision-browser-q4 with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull nerozx/qwen3.5-2b-decision-browser-q4:Q8_0
Run and chat with the model
lemonade run user.qwen3.5-2b-decision-browser-q4-Q8_0
List all available models
lemonade list
- Hermes Agent
How to use nerozx/qwen3.5-2b-decision-browser-q4 with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf nerozx/qwen3.5-2b-decision-browser-q4:Q8_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://127.0.0.1:9931/v1 hermes config set model.default nerozx/qwen3.5-2b-decision-browser-q4:Q8_0
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use nerozx/qwen3.5-2b-decision-browser-q4 with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf nerozx/qwen3.5-2b-decision-browser-q4:Q8_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://127.0.0.1:9931/v1 \ --custom-model-id "nerozx/qwen3.5-2b-decision-browser-q4:Q8_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"
Qwen3.5-2B ยท Unsloth Decision head (Clef) โ project-browser, Q4_K_M
A decision model, not a chat model. Given a state (task + step + previous
actions) and a set of questions whose options are page elements, it returns the
option key to interact with next, plus a calibrated probability per option.
Served by Unsloth Studio's /v1/systemone (llama.cpp backend, clef layout).
Files
| file | size | note |
|---|---|---|
model-Q4_K_M.gguf |
1.29 GB | full checkpoint: backbone + joint decision head |
mmproj-Q8_0.gguf |
365 MB | vision tower (only needed for image input) |
export.json |
โ | Studio export contract (layout, quantization, fingerprint) |
unsloth_decision_config.json |
โ | temperatures, training metadata |
joint_head_config.json |
โ | head shape (hidden 2048, width 512, 4 layers) |
Do not run this with a chat prompt. It needs a state + questions request.
Request shape
POST /v1/systemone
{
"model": "default",
"state": "Task: click the Interfaces row in the Health table\nWebsite: example.com\nStep: 4 of 8\nPrevious actions:\n(none)",
"questions": {
"target": {
"type": "choice",
"instructions": "Which element on the page should be interacted with next ...",
"criteria": {"e00": "<div role=\"row\"> Interfaces 86 50 1", "e01": "...", "none": "None of the listed elements ..."}
},
"operation": {"type": "choice", "instructions": "What should be done ...",
"criteria": {"click": "CLICK ...", "type": "TYPE ...", "select": "SELECT ..."}},
"needs_value": {"type": "noul", "instructions": "Does this action need a specific value ..."}
}
}
Response: answers.target.choice + confidence + probabilities (calibrated).
Envelope (measured, do not exceed)
- โค10 options per question (+
none). The calibration temperatures are fit per option-count bucket (โค2 / 3โ5 / 6โ10 / 11+); at 41โ53 options the distribution becomes uninformative (flat or sharp-but-wrong). - Atomic instruction (one step, English). Multi-step SOPs are the planner's job, not this head's.
Website:= hostname only. A full URL leaks tenant/slug tokens that anchor the head to unrelated elements.
Evaluation (eval split, N=375, target accuracy)
| quantization | accuracy | ECE |
|---|---|---|
| Q8_0 / MLX fp16 (source) | 88.3% | 0.0506 |
| Q4_K_M (this file) | 86.9% | 0.0422 |
Agreement Q8 vs Q4: 96.3%. none recall 100% in both; no false abstentions.
The joint head is quantized too (Q4_K/Q6_K weights, F32 norms/bias) โ the
measured cost is 1.4 pp accuracy with slightly better calibration.
Intended use / limits
For: picking the next element in a browser agent loop, one step at a time, with a pre-filtered candidate list (a ranker or a planner supplies โค10 options).
Not for: free-text generation, writing field values, judging whether a task is complete, planning multi-step procedures, or searching a large DOM (it does not rank hundreds of elements โ pass it a shortlist).
Provenance
- Base:
unsloth/Qwen3.5-2B(QLoRA,soft_cross_entropy, clef head). - SFT data: Mind2Web-derived decision rows (train/eval splits, ~6.9k steps). Mind2Web is CC-BY-4.0; derivative data keeps that attribution.
- Exported from an Unsloth Studio decision checkpoint; see
export.jsonsource_fingerprintfor the exact source folder state.
- Downloads last month
- 110
4-bit