# Gemma 4 静默更新：修复工具调用错误并提升 Hopper GPU 性能

- 来源：The Decoder：AI News（RSS）
- 作者：Jonathan Kemper
- 发布时间：2026-07-16 17:07
- AIHOT 分数：48
- AIHOT 链接：https://aihot.virxact.com/items/cmrnb8azf00t8bic685er9bs3
- 原文链接：https://the-decoder.com/gemma-4-gets-a-stealth-update-that-fixes-tool-calling-bugs-and-truncated-responses-under-the-same-name

## AI 摘要

Google 为开源模型 Gemma 4 推送更新，在 Nvidia Hopper GPU 上启用 Flash Attention 4 后，提示词处理速度提升 25-70%，首 token 延迟降低 31%。更新还修复了工具调用错误和响应截断问题，Gemma 4 31B 在电信场景的智能体推理性能提升 10.1%。

## 正文

Google shipped an update to its open AI model Gemma 4 that speeds up performance on Nvidia Hopper GPUs, fixes tool calling bugs, and addresses problems with truncated responses. Turning on Flash Attention 4 boosts the speed at which the model processes incoming prompts by 25 to 70 percent, according to Google. Time to first token drops by up to 31 percent. Google also fixed bugs in tool calling, the feature that lets the model trigger external tools on its own.

Gemma 4 31B improves its agentic reasoning and tool-calling performance across all tested scenarios, with the highest increase of 10.1% in the telecommunications use case.

Google says it also cut down on cases where the model would cut answers short or return incomplete responses. For image processing, users can manually raise the "max_soft_tokens" parameter from 280 to 1,120 to get sharper OCR results and support resolutions up to 2.51 megapixels. Google put up an interactive configurator on Hugging Face for that. The published benchmarks only compare the 31B and E4B variants against their predecessors, but the Hugging Face repository shows that all parameter sizes in this model generation got updated, including the newest 12B release. The community has pushed back on Google shipping the update under the same "Gemma 4" name instead of tagging it as a separate version like "Gemma 4.1."