Text Generation
MLX
Safetensors
English
qwen3_5
bonsai
oqe
calibration-smoke
experimental
not-for-production
conversational
4-bit precision
Instructions to use TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke") prompt = "Write a story about Einstein" messages = [{"role": "user", "content": prompt}] prompt = tokenizer.apply_chat_template( messages, add_generation_prompt=True ) text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Pi
How to use TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke with Pi:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke"
Configure the model in Pi
# Install Pi: npm install -g @mariozechner/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "mlx-lm": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke with Hermes Agent:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke"
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 TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke
Run Hermes
hermes
- OpenClaw new
How to use TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke with OpenClaw:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke"
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 "TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke" \ --custom-provider-id mlx-lm \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
- MLX LM
How to use TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Interactive chat REPL mlx_lm.chat --model "TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke"
Run an OpenAI-compatible server
# Install MLX LM uv tool install mlx-lm # Start the server mlx_lm.server --model "TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke" # Calling the OpenAI-compatible server with curl curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "TiGa-RCE/Bonsai-27B-oQ4e-S32-Smoke", "messages": [ {"role": "user", "content": "Hello"} ] }'
| { | |
| "model_id": "oq4e-smoke-s32", | |
| "benchmark": "arc_challenge", | |
| "sampling": "deterministic_thinking_off", | |
| "total": 10, | |
| "correct": 6, | |
| "accuracy": 0.6, | |
| "time_s": 39.548, | |
| "peak_memory_gb": 15.415, | |
| "question_results": [ | |
| { | |
| "id": "Mercury_7043820", | |
| "expected": "B", | |
| "predicted": "A", | |
| "correct": false, | |
| "raw_response": "A." | |
| }, | |
| { | |
| "id": "MCAS_2003_5_33", | |
| "expected": "C", | |
| "predicted": "C", | |
| "correct": true, | |
| "raw_response": "C" | |
| }, | |
| { | |
| "id": "Mercury_7219678", | |
| "expected": "A", | |
| "predicted": "B", | |
| "correct": false, | |
| "raw_response": "The question describes a scenario where a beach community's water supply (surface and groundwater) is decreasing. They decide to build a large-scale desalination plant to remove salt from bay water. The key reason for this decision is the **decrease in water resources**.\n\nLet's evaluate the options:\n\nA. **needed increased resources.** - This aligns with the problem statement: the water supply is decreasing, so they need more water resources. Desalination is a method to create fresh water resources from saltwater.\nB. **had access to advanced technology.** - While desalination uses technology, the *reason* for" | |
| }, | |
| { | |
| "id": "AKDE&ED_2008_8_48", | |
| "expected": "A", | |
| "predicted": "A", | |
| "correct": true, | |
| "raw_response": "A" | |
| }, | |
| { | |
| "id": "Mercury_7210455", | |
| "expected": "C", | |
| "predicted": "C", | |
| "correct": true, | |
| "raw_response": "C" | |
| }, | |
| { | |
| "id": "Mercury_407517", | |
| "expected": "C", | |
| "predicted": "C", | |
| "correct": true, | |
| "raw_response": "C" | |
| }, | |
| { | |
| "id": "TIMSS_2007_8_pg109", | |
| "expected": "C", | |
| "predicted": "C", | |
| "correct": true, | |
| "raw_response": "C" | |
| }, | |
| { | |
| "id": "Mercury_SC_400134", | |
| "expected": "D", | |
| "predicted": "B", | |
| "correct": false, | |
| "raw_response": "B" | |
| }, | |
| { | |
| "id": "Mercury_402501", | |
| "expected": "B", | |
| "predicted": "B", | |
| "correct": true, | |
| "raw_response": "B" | |
| }, | |
| { | |
| "id": "Mercury_400084", | |
| "expected": "D", | |
| "predicted": "C", | |
| "correct": false, | |
| "raw_response": "To balance the chemical equation for the combustion of methane:\n\n$$ \\text{CH}_4 + \\Box \\text{O}_2 \\rightarrow 2\\text{CO}_2 + 4\\text{H}_2\\text{O} $$\n\nLet's count the atoms on each side:\n\n**Right Side (Products):**\n* Carbon (C): 2 atoms (from $2\\text{CO}_2$)\n* Hydrogen (H): 8 atoms (from $4\\text{H}_2\\text{O}$)\n* Oxygen (O): $(2" | |
| } | |
| ] | |
| } | |