Ready-to-use verification hashes for Transformers.js and WebLLM models. Skip the hash generation step and start verifying immediately.

Install

npm install @verifyfetch/manifests

Quick Start

Import a manifest directly and pass it to any verifyfetch integration:

import { verifiedPipeline } from '@verifyfetch/transformers';
import manifest from '@verifyfetch/manifests/transformers/Xenova--distilbert-base-uncased-finetuned-sst-2-english.json';

const classifier = await verifiedPipeline(
  'sentiment-analysis',
  'Xenova/distilbert-base-uncased-finetuned-sst-2-english',
  { manifest }
);

Available Models

Transformers.js

| Model | Import | |-------|--------| | Xenova/distilbert-base-uncased-finetuned-sst-2-english | @verifyfetch/manifests/transformers/Xenova--distilbert-base-uncased-finetuned-sst-2-english.json | | Xenova/all-MiniLM-L6-v2 | @verifyfetch/manifests/transformers/Xenova--all-MiniLM-L6-v2.json |

WebLLM

| Model | Import | |-------|--------| | Phi-3-mini-4k-instruct-q4f16_1-MLC | @verifyfetch/manifests/webllm/Phi-3-mini-4k-instruct-q4f16_1-MLC.json |

API

import { availableModels, getManifestPath } from '@verifyfetch/manifests';

// List available models
console.log(availableModels.transformers);
// ['Xenova/distilbert-base-uncased-finetuned-sst-2-english', 'Xenova/all-MiniLM-L6-v2']

console.log(availableModels.webllm);
// ['Phi-3-mini-4k-instruct-q4f16_1-MLC']

// Get import path for a model
const path = getManifestPath('transformers', 'Xenova/all-MiniLM-L6-v2');
// '@verifyfetch/manifests/transformers/Xenova--all-MiniLM-L6-v2.json'

Generate Your Own

To generate manifests for models not included here:

npx @verifyfetch/cli hash-model <model-id>

Related

• verifyfetch - Core library
• @verifyfetch/transformers - Transformers.js integration
• @verifyfetch/webllm - WebLLM integration
• GitHub

License

Apache-2.0