Instructions to use daveycodez/LFM3-1.7B-Web with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers.js
How to use daveycodez/LFM3-1.7B-Web with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('text-generation', 'daveycodez/LFM3-1.7B-Web');
Commit History
Today
-
Run the sliding-window layers with flash attention
Verified signature3c75909 -
Block-32 GPTQ with a 48k vocabulary (900 MB of weights)
Verified signature2395323 -
Trim the vocabulary to 80k tokens (856 MB of weights)
Verified signaturec7cce6b -
Remove the Reshape that duplicated the embedding at load
Verified signaturea405c7f -
LFM3-1.7B ONNX, GPTQ 4-bit (block 128), 16k rotary caches
Verified signature5bc567e -
initial commit
Verified signaturef7e20d4