Text Generation Inference v3.3.7 发布：新增图像大小限制与维护模式

AI 摘要

Hugging Face 发布了文本生成推理库 Text Generation Inference 的 v3.3.7 版本。该版本新增了 max_image_fetch_size 参数以限制图像获取大小，修复了当 NVIDIA_VISIBLE_DEVICES 为 'all' 或 'void' 时设备数量自动计算的问题，并引入了维护模式。这些更新旨在提升推理服务的稳定性和配置灵活性。

正文节选

## What's Changed * misc(gha): expose action cache url and runtime as secrets by @mfuntowicz in https://github.com/huggingface/text-generation-inference/pull/2964 * feat: support max_image_fetch_size to limit by @drbh in https://github.com/huggingface/text-generation-inference/pull/3339 * Maintenance mode by @LysandreJik in https://github.com/huggingface/text-generation-inference/pull/3344 * Maintenance mode by @LysandreJik in https://github.com/huggingface/text-generation-inference/pull/334

发布时间：2025-12-19 22:35

抓取时间：2026-08-02 00:31

来源机构：Hugging Face