Instructions to use ZYLIM/qwen3.5-0.8b-deapi-shards with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use ZYLIM/qwen3.5-0.8b-deapi-shards with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir qwen3.5-0.8b-deapi-shards ZYLIM/qwen3.5-0.8b-deapi-shards
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download manifest.json from ZYLIM/qwen3.5-0.8b-deapi-shards: direct link, hf CLI and curl.
- Browser
- Download file 685 Bytes
-
https://huggingface.co/ZYLIM/qwen3.5-0.8b-deapi-shards/resolve/main/manifest.json
- Command line
-
hf download hf://ZYLIM/qwen3.5-0.8b-deapi-shards/manifest.json
-
curl -L -o manifest.json https://huggingface.co/ZYLIM/qwen3.5-0.8b-deapi-shards/resolve/main/manifest.json
685 Bytes
| { | |
| "format": "deapi-model-shard", | |
| "version": 1, | |
| "model": "qwen3.5-0.8b", | |
| "architecture": "qwen3.5", | |
| "num_nodes": 2, | |
| "num_layers": 24, | |
| "shards": [ | |
| { | |
| "node_id": 0, | |
| "filename": "qwen3.5-0.8b-node0.dms", | |
| "size_bytes": 1023952896, | |
| "layer_start": 0, | |
| "layer_end": 12, | |
| "has_embed": true, | |
| "has_head": false, | |
| "has_tokenizer": true, | |
| "num_tensors": 160 | |
| }, | |
| { | |
| "node_id": 1, | |
| "filename": "qwen3.5-0.8b-node1.dms", | |
| "size_bytes": 1006699008, | |
| "layer_start": 12, | |
| "layer_end": 24, | |
| "has_embed": false, | |
| "has_head": true, | |
| "has_tokenizer": false, | |
| "num_tensors": 161 | |
| } | |
| ] | |
| } |