Text Generation
MLX
Safetensors
English
qwen3-mdlm-adaptive-hybrid
qwen3
diffusion
text-diffusion
chat
Instructions to use goldenfox/marimo-0.6b-mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use goldenfox/marimo-0.6b-mlx with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # if on a CUDA device, also pip install mlx[cuda] # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("goldenfox/marimo-0.6b-mlx") prompt = "Once upon a time in" text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- MLX LM
How to use goldenfox/marimo-0.6b-mlx with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Generate some text mlx_lm.generate --model "goldenfox/marimo-0.6b-mlx" --prompt "Once upon a time"
- Atomic Chat
| { | |
| "model_type": "qwen3-mdlm-adaptive-hybrid", | |
| "vocab_size": 151936, | |
| "hidden_size": 1024, | |
| "num_hidden_layers": 28, | |
| "num_attention_heads": 16, | |
| "num_key_value_heads": 8, | |
| "head_dim": 128, | |
| "intermediate_size": 3072, | |
| "rms_norm_eps": 1e-06, | |
| "rope_theta": 1000000.0, | |
| "tie_word_embeddings": true, | |
| "mdlm": { | |
| "mask_token_id": 151672, | |
| "think_id": 151667, | |
| "end_think_id": 151668, | |
| "thought_pad_id": 151673, | |
| "size_ids": { | |
| "32": 151674, | |
| "64": 151675, | |
| "128": 151676 | |
| }, | |
| "eos_id": 151643, | |
| "im_end_id": 151645, | |
| "forbidden_output_token_ids": [ | |
| 151667, | |
| 151669, | |
| 151670, | |
| 151671, | |
| 151672 | |
| ], | |
| "forbidden_output_from": 151679, | |
| "max_seq_len": 2048, | |
| "causal_prefix": true, | |
| "steps_per_block": 16, | |
| "max_answer_tokens": 384, | |
| "keep_messages": 4, | |
| "system_prompt": "You are a claim intake assistant. Track every detail the customer gives and always use the corrected value when they correct themselves.", | |
| "checkpoint_step": 3999 | |
| } | |
| } | |