ONNX export for browser/local inference. Converted from Qwen/Qwen2.5-0.5B-Instruct at revision 7ae557604adf67be50417f59c2c2f167def9a775. No fine-tuning. — Task: text-generation-with-past — Precision: fp32, fp16 — Context: 32768 tokens — Runtime: Transformers.js / ONNX Runtime; select the dtype explicitly — Files: weights in onnx/; tokenizer and configuration at the repository root — Provenance and hashes: build-manifest.json, SHA256SUMS FP32 requires both onnx/model.onnx and onnx/model.onnxdata. FP16 uses onnx/modelfp16.onnx. INT8 and Q4 weights are not included. Validation reports and helper scripts are in validation/ and tools/. See the upstream model card for model details and licensing.
Модальности:
Генерация текста
Области применения:
Диалог / чат Следование инструкциям
Задача: Генерация текста
Автор: JasonYANG170
Теги: transformers.js, onnx, qwen2, webgpu, conversational, zh
Лайков: 3 | Загрузок: 608
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
