Instructions to use DownFlow/webasmr-llm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers.js
How to use DownFlow/webasmr-llm with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('text-generation', 'DownFlow/webasmr-llm');
Download tokenizer.json from DownFlow/webasmr-llm: direct link, hf CLI and curl.
- Browser
- Download file 19.2 MB
-
https://huggingface.co/DownFlow/webasmr-llm/resolve/main/tokenizer.json
- Command line
-
hf download hf://DownFlow/webasmr-llm/tokenizer.json
-
curl -L -o tokenizer.json https://huggingface.co/DownFlow/webasmr-llm/resolve/main/tokenizer.json
19.2 MB
- Xet hash:
- 8b5fc71116be0275aa12efae0db35e92e9227f80264f569bbe283ef170f1bbc7
- Size of remote file:
- 19.2 MB
- SHA256:
- 89da80cc6689bef4d90cc1028249436975ffb0814618f1d93c65310e05801a9b
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.