English(EN) Zhipu is testing a faster version of its GLM-5.3 model, with users reporting roughly 490 tokens per second - more than double the 200 tokens/s of the FlashX var

智谱AI测试GLM-5.3模型，速度达每秒490个token

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-13 02:34

智谱AI正在测试其GLM-5.3大语言模型的加速版本，据报道速度约为每秒490个token。这款新的“Highspeed”变体比之前的“FlashX”变体快一倍多，后者每秒处理约200个token。更快的模型目前正在分发给部分ZCode订阅用户。 AI

影响这种速度的提升可以显著增强实时AI应用以及用户与智谱模型的交互。

排序理由 Frontier-lab模型发布，带有系统卡[lever_c_demoted from frontier_release: ic=1 ai=1.0]

在 Mastodon — mastodon.social 阅读 →

模型发布

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 10 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Significant; Frontier-lab模型发布，带有系统卡[lever_c_demoted from frontier_release: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

Mastodon — mastodon.social TIER_1 English(EN) · [email protected] · 2026-10-13 02:34

智谱正在测试其GLM-5.3模型的更快版本，用户报告称每秒约490个token——是FlashX var的200 token/s的两倍多

Zhipu is testing a faster version of its GLM-5.3 model, with users reporting roughly 490 tokens per second - more than double the 200 tokens/s of the FlashX variant. The Highspeed variant is being rolled out to randomly selected ZCode subscribers. https:// pandaily.com/zhipu-glm-…

链接 pandaily.com/zhipu-glm-5-3-highspeed-zcod…

本文如何被排名

报道来源 [1]

智谱正在测试其GLM-5.3模型的更快版本，用户报告称每秒约490个token——是FlashX var的200 token/s的两倍多

相关实体

相关话题