Image-Text-to-Text
MLX
Safetensors
inkling2_loop
quantized
Mixture of Experts
conversational
4-bit precision
Instructions to use araysuter/Inkling-2-Mini-MLX-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use araysuter/Inkling-2-Mini-MLX-4bit with MLX:
# Make sure mlx-vlm is installed # pip install --upgrade mlx-vlm from mlx_vlm import load, generate from mlx_vlm.prompt_utils import apply_chat_template from mlx_vlm.utils import load_config # Load the model model, processor = load("araysuter/Inkling-2-Mini-MLX-4bit") config = load_config("araysuter/Inkling-2-Mini-MLX-4bit") # Prepare input image = ["http://images.cocodataset.org/val2017/000000039769.jpg"] prompt = "Describe this image." # Apply chat template formatted_prompt = apply_chat_template( processor, config, prompt, num_images=1 ) # Generate output output = generate(model, processor, formatted_prompt, image) print(output) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Pi
How to use araysuter/Inkling-2-Mini-MLX-4bit with Pi:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "araysuter/Inkling-2-Mini-MLX-4bit"
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "mlx-lm": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "araysuter/Inkling-2-Mini-MLX-4bit" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent
How to use araysuter/Inkling-2-Mini-MLX-4bit 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 "araysuter/Inkling-2-Mini-MLX-4bit"
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 araysuter/Inkling-2-Mini-MLX-4bit
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use araysuter/Inkling-2-Mini-MLX-4bit with OpenClaw:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "araysuter/Inkling-2-Mini-MLX-4bit"
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 "araysuter/Inkling-2-Mini-MLX-4bit" \ --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"
Download VALIDATION.json from araysuter/Inkling-2-Mini-MLX-4bit: direct link, hf CLI and curl.
- Browser
- Download file 5.01 kB
-
https://huggingface.co/araysuter/Inkling-2-Mini-MLX-4bit/resolve/main/VALIDATION.json
- Command line
-
hf download hf://araysuter/Inkling-2-Mini-MLX-4bit/VALIDATION.json
-
curl -L -o VALIDATION.json https://huggingface.co/araysuter/Inkling-2-Mini-MLX-4bit/resolve/main/VALIDATION.json
5.01 kB
| { | |
| "source": { | |
| "profile": "source", | |
| "load_seconds": 2.8372250832617283, | |
| "source_revision": "6e7c8248f75265bdd4fbab5273304243b5d8c2cb", | |
| "runtime_commit": "0c4c99192230f1943dd468d297f0b8a64ae7b7c2", | |
| "cases": [ | |
| { | |
| "name": "arithmetic", | |
| "output": "391", | |
| "tokens": 58, | |
| "elapsed_s": 3.820709999650717, | |
| "prompt_tps": 85.20184649445878, | |
| "generation_tps": 22.839770376876665, | |
| "peak_memory_gb": 19.678987508, | |
| "token_sha256": "f477f731aca97557510c017ad282b5ec1936fef6da325f8a8945125f388f7cea", | |
| "available_gib": 14.812255859375 | |
| }, | |
| { | |
| "name": "coding", | |
| "output": "```python\ndef larger(a, b):\n return max(a, b)\n```", | |
| "tokens": 78, | |
| "elapsed_s": 3.671069875359535, | |
| "prompt_tps": 101.70025005118325, | |
| "generation_tps": 23.06434204113524, | |
| "peak_memory_gb": 19.68460474, | |
| "token_sha256": "46508297f2a596be317928d87a58bbe8b7817d1a529910bcc13eeaca3b421f62", | |
| "available_gib": 14.664840698242188 | |
| }, | |
| { | |
| "name": "vision", | |
| "output": "The image shows a red square on the left and a blue circle on the right.", | |
| "tokens": 78, | |
| "elapsed_s": 3.7814209582284093, | |
| "prompt_tps": 60.030500801066125, | |
| "generation_tps": 26.640466036111427, | |
| "peak_memory_gb": 19.768098972, | |
| "token_sha256": "a39ce756075e66440c0b32ebf2843fd21cd10fd7ce6fceb8256444c7756868bd", | |
| "available_gib": 14.058120727539062 | |
| } | |
| ], | |
| "layer_visits": 48, | |
| "public_text_probe": { | |
| "text": "A scientist compares two measurements by subtracting the smaller value from the larger value. A careful software engineer writes a small test before changing a calculation. The list contains three colors: red, green, and blue. A dictionary stores keys and values. Water freezes at zero degrees Celsius under ordinary atmospheric pressure. To find the area of a rectangle, multiply its length by its width. Good documentation explains the assumptions, the inputs, and the expected outputs.", | |
| "tokens": 90, | |
| "token_ids_sha256": "6b76f6dc31408df063308af8b212e38c5dadf8d262feb94f5899f4dc7e298a50", | |
| "nll": 2.266077995300293, | |
| "perplexity": 9.641512506217301 | |
| }, | |
| "status": "passed" | |
| }, | |
| "converted": { | |
| "profile": "Inkling-2-Mini-MLX-4bit", | |
| "load_seconds": 1.2451343750581145, | |
| "source_revision": "6e7c8248f75265bdd4fbab5273304243b5d8c2cb", | |
| "runtime_commit": "0c4c99192230f1943dd468d297f0b8a64ae7b7c2", | |
| "cases": [ | |
| { | |
| "name": "arithmetic", | |
| "output": "391", | |
| "tokens": 48, | |
| "elapsed_s": 1.20463312510401, | |
| "prompt_tps": 157.64685810416975, | |
| "generation_tps": 50.58565854919863, | |
| "peak_memory_gb": 17.618073116, | |
| "token_sha256": "9020bf4267f9055490be88ab82311e017c1b4223135d13c18336e5783ee52c34", | |
| "available_gib": 34.53157043457031 | |
| }, | |
| { | |
| "name": "coding", | |
| "output": "```python\ndef larger(a, b):\n return max(a, b)\n```", | |
| "tokens": 103, | |
| "elapsed_s": 2.203887917101383, | |
| "prompt_tps": 211.51820034274786, | |
| "generation_tps": 49.89645479238404, | |
| "peak_memory_gb": 17.627712916, | |
| "token_sha256": "f7d774b67bb547f788aa4032480022f036d13de76bfcbacf8acb42d7691d389c", | |
| "available_gib": 34.524658203125 | |
| }, | |
| { | |
| "name": "vision", | |
| "output": "The image shows a red square on the left and a blue circle on the right.", | |
| "tokens": 75, | |
| "elapsed_s": 1.700042124837637, | |
| "prompt_tps": 254.43959687493455, | |
| "generation_tps": 50.09135688078983, | |
| "peak_memory_gb": 17.705621068, | |
| "token_sha256": "bf31bc849321ced525902c3745e58e9a6710405377be01f0b63d4308c90913e8", | |
| "available_gib": 34.52320861816406 | |
| } | |
| ], | |
| "layer_visits": 48, | |
| "public_text_probe": { | |
| "text": "A scientist compares two measurements by subtracting the smaller value from the larger value. A careful software engineer writes a small test before changing a calculation. The list contains three colors: red, green, and blue. A dictionary stores keys and values. Water freezes at zero degrees Celsius under ordinary atmospheric pressure. To find the area of a rectangle, multiply its length by its width. Good documentation explains the assumptions, the inputs, and the expected outputs.", | |
| "tokens": 90, | |
| "token_ids_sha256": "6b76f6dc31408df063308af8b212e38c5dadf8d262feb94f5899f4dc7e298a50", | |
| "nll": 2.2467293739318848, | |
| "perplexity": 9.456755690057983 | |
| }, | |
| "status": "passed" | |
| }, | |
| "smoke_checks": { | |
| "arithmetic_391": true, | |
| "python_function_parses": true, | |
| "vision_colors_and_shapes": true | |
| }, | |
| "text_probe_delta_nll": -0.019348621368408203, | |
| "text_probe_perplexity_ratio": 0.9808373617686875, | |
| "benchmark_caveat": "Tiny smoke test and one synthetic text probe; not a standard quality benchmark or a full context qualification." | |
| } | |