𝕏xCN first seen 1 d ago, last 27 min ago, peak #1

Zhipu rolls out GLM-5.3-Highspeed hitting 490 tokens per second

Original: BREAKING: Zhipu appears to have a new GLM model hitting 490 tokens/s. GLM-5.3-Highspeed runs more than 2× faster than GL

Chinese AI firm Zhipu appears to have launched a new fast-inference model, GLM-5.3-Highspeed, reportedly reaching 490 tokens per second — more than double the roughly 200 tokens/s of GLM-5.3-FlashX. The model is being rolled out gradually in ZCode 3.15.1, with access randomly available to some users so far. AI watchers are tracking the release for confirmation and benchmarks.

Why now: A major Chinese AI lab may have quietly released a dramatically faster model, prompting users to test access and verify claims.

ZhipuGLM-5.3-HighspeedGLM-5.3-FlashXZCode

Open on x →

Rank over time, top of the chart is #1. 2 snapshots from 1 d ago to 27 min ago.

Evidence

• BREAKING: Zhipu appears to have a new GLM model hitting 490 tokens/s. GLM-5.3-Highspeed runs more than 2× faster than GLM-5.3-FlashX, which peaks at around 200 tokens/s. The model is currently rolling out gradually in ZCode 3.15.1, with access randomly available to · Chinazhidx · 231

API: https://socialmediatrends-api.osmike.com/v1/trends/1945665