Instructions to use Meddies/lfm2.5-230m-title-onnx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers.js
How to use Meddies/lfm2.5-230m-title-onnx with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('text-generation', 'Meddies/lfm2.5-230m-title-onnx');
LFM2.5-230M ONNX q4, Meddies mirror for on-device chat titles
Unmodified mirror of the q4 ONNX variant of LiquidAI/LFM2.5-230M-ONNX
(source revision c6f46e4e3f885ebcad164d14059a49f90e27eb4d), pinned by sha256 in the Meddies model registry. It generates the sidebar title of a
conversation in the doctor's browser on WebGPU; the query never leaves the device for this purpose.
Files: onnx/model_q4.onnx + onnx/model_q4.onnx_data, tokenizer.json, tokenizer_config.json, config.json,
generation_config.json, chat_template.jinja, LICENSE (LFM Open License v1.0, verbatim), NOTICE, manifest.json
(bytes + sha256 per file).
Fine-tuned title weights will be published to this same repository as a new revision; the registry row pins the revision.
- Downloads last month
- 213
Model tree for Meddies/lfm2.5-230m-title-onnx
Base model
LiquidAI/LFM2.5-230M-Base Finetuned
LiquidAI/LFM2.5-230M Quantized
LiquidAI/LFM2.5-230M-ONNX