Text Generation
MLX
Safetensors
Polish
English
llama
typed-decisions
decision-making
classification
calibration
polish
6-bit
conversational
Instructions to use agentGreg/basal-1.5-mini-MLX-6bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use agentGreg/basal-1.5-mini-MLX-6bit 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("agentGreg/basal-1.5-mini-MLX-6bit") 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
- MLX LM
How to use agentGreg/basal-1.5-mini-MLX-6bit with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Interactive chat REPL mlx_lm.chat --model "agentGreg/basal-1.5-mini-MLX-6bit"
Run an OpenAI-compatible server
# Install MLX LM uv tool install mlx-lm # Start the server mlx_lm.server --model "agentGreg/basal-1.5-mini-MLX-6bit" # Calling the OpenAI-compatible server with curl curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "agentGreg/basal-1.5-mini-MLX-6bit", "messages": [ {"role": "user", "content": "Hello"} ] }' - Atomic Chat
Download CALIBRATION.json from agentGreg/basal-1.5-mini-MLX-6bit: direct link, hf CLI and curl.
- Browser
- Download file 1.12 kB
-
https://huggingface.co/agentGreg/basal-1.5-mini-MLX-6bit/resolve/main/CALIBRATION.json
- Command line
-
hf download hf://agentGreg/basal-1.5-mini-MLX-6bit/CALIBRATION.json
-
curl -L -o CALIBRATION.json https://huggingface.co/agentGreg/basal-1.5-mini-MLX-6bit/resolve/main/CALIBRATION.json
1.12 kB
| { | |
| "temperature_per_prim": { | |
| "choice": 1.1387886347566916, | |
| "noul": 1.3939382634489992, | |
| "score": 1.5422108254079405 | |
| }, | |
| "thresholds": { | |
| "0.01": { | |
| "confidence": 0.9546241492914076, | |
| "method": "bound-controlled (one-sided 95% Clopper-Pearson on FIT)", | |
| "cert_coverage": 0.49265381083562904, | |
| "cert_error": 0.009319664492078328, | |
| "cert_upper95": 0.015756935615729575, | |
| "test_coverage": 0.4649532710280374, | |
| "test_error": 0.009045226130653283, | |
| "test_pl_coverage": 0.5283152097161418, | |
| "test_pl_error": 0.007751937984496138 | |
| }, | |
| "0.05": { | |
| "confidence": 0.7629656203142733, | |
| "method": "bound-controlled (one-sided 95% Clopper-Pearson on FIT)", | |
| "cert_coverage": 0.7112029384756657, | |
| "cert_error": 0.046481601032924424, | |
| "cert_upper95": 0.05626481588712098, | |
| "test_coverage": 0.704088785046729, | |
| "test_error": 0.05292848846855813, | |
| "test_pl_coverage": 0.7709363084310126, | |
| "test_pl_error": 0.0426818098552848 | |
| } | |
| }, | |
| "fitted_on": "calib FIT split (70% by twin group)", | |
| "certified_on": "calib CERT split (30%)", | |
| "evaluated_on": "test (frozen)", | |
| "eval_dtype": "float32" | |
| } |