把 ONNX、PyTorch、Keras、CoreML 等格式的神经网络模型直接拖进浏览器就能看结构图，排查模型转换和层连接问题很方便。
Rust 写的张量库和深度学习框架，兼顾灵活性与效率，支持自动微分、跨平台后端和内核融合，可训练和部署神经网络。
safetensors 的扩展，能把稀疏量化后的张量高效存到磁盘，省空间又不牺牲加载速度，适合大模型量化权重的持久化。
…oaded. - Klein 9B's VAE is now saved as flux2-vae.safetensors , the same file Flux 3 and Fibo 2.0 use. It used to be saved as ae.safetensors , the name of Z-Image's differ
… it also holds the converted codec link, codec-v1.safetensors . If it was deleted, setup-voice.sh converts the codec again in a few seconds. 🤖 Generated with [Claude Code
…m/MinishLab/model2vec/pull/338 fix: guard against safetensors contiguity bug by @stephantul in https://github.com/MinishLab/model2vec/pull/341 feat: bump skeletoken & upda
…on, and recipe workflows. Highlights: - use exact safetensors parameter counts when available - account for stored and active experts in MoE models - separate quantization
…rch aliases - FIx vLLM v1 issues - Extend save to safetensors to VLMs Full Changelog : https://github.com/mobiusml/hqq/compare/v0.2.5...0.2.6
…y ( 733) by @zly-idleness - Migrate all models to safetensors (b9b007ef) by @fumiama --- 新增 - NPU 支持 ( 777) by @shen-shanshan 修复 - 中文正则化器中缺少 remove interjections ( 783) by