Instructions to use Intel/Cosmos3-Nano-int4-AutoRound with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Intel/Cosmos3-Nano-int4-AutoRound with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Intel/Cosmos3-Nano-int4-AutoRound", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Download transformer/quantization_config.json from Intel/Cosmos3-Nano-int4-AutoRound: direct link, hf CLI and curl.
- Browser
- Download file 343 Bytes
-
https://huggingface.co/Intel/Cosmos3-Nano-int4-AutoRound/resolve/main/transformer/quantization_config.json
- Command line
-
hf download hf://Intel/Cosmos3-Nano-int4-AutoRound/transformer/quantization_config.json
-
curl -L -o quantization_config.json https://huggingface.co/Intel/Cosmos3-Nano-int4-AutoRound/resolve/main/transformer/quantization_config.json
343 Bytes
| { | |
| "autoround_version": "0.14.0", | |
| "batch_size": 1, | |
| "bits": 4, | |
| "block_name_to_quantize": [ | |
| "language_model.layers", | |
| "gen_layers", | |
| "blocks", | |
| "deepstack_merger_list" | |
| ], | |
| "data_type": "int", | |
| "group_size": 128, | |
| "nsamples": 16, | |
| "packing_format": "auto_round:auto_gptq", | |
| "quant_method": "auto-round", | |
| "sym": true | |
| } |