VQV Signal

OPEN SOURCE WATCH TECHNICAL

Baidu releases ERNIE 5.5 Flash Preview and promises open weights by November 25

According to a post on X by @Baidu_Inc, Baidu released ERNIE 5.5 Flash Preview, a mixture-of-experts model with 280B total and 14B active parameters, a 1M-token context and more than 200 tokens per second. The post says the open weights will be released by November 25. A related post from @Baidu_ERNIE says it is the first post-trained model built on ERNIE 5.5 Flash Base. Commentary in the same feed from @CobusGreylingZA expects agent builders to wait until the weights actually ship.

Source: Scobleizer News Pulse (X) · x.com Found via Robert Scoble’s Scobleizer News Pulse · a post on X, not independently verified Published 2026-11-11T14:55:00+00:00 Detected 2026-11-11T22:20:18+00:00
View original source

According to a post on X by @Baidu_Inc, Baidu released ERNIE 5.5 Flash Preview, a mixture-of-experts model with 280B total and 14B active parameters, a 1M-token context and more than 200 tokens per second. The post says the open weights will be released by November 25. A related post from @Baidu_ERNIE says it is the first post-trained model built on ERNIE 5.5 Flash Base. Commentary in the same feed from @CobusGreylingZA expects agent builders to wait until the weights actually ship.

AI-assisted summary based on the listed source.

Baidu released ERNIE 5.5 Flash Preview, a mixture-of-experts model with 280B total and 14B active parameters, a 1M-token context and more than 200 tokens per second, and says the open weights will be released by November 25. @Baidu_ERNIE says it is the first post-trained model on the ERNIE 5.5 Flash Base....

Developers and businesses weighing open-weight models for agent work now have a specific date to track, but the weights have been promised rather than released. Builders may prefer to wait until the weights are published before committing, as one commenter in the feed suggests. Watch for the November 25 release and whether the published weights match the stated specifications.

Signal Strength 91% Technical label WATCH Public Interest 37 Category OPEN SOURCE Reader Depth TECHNICAL

Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.

Public Interest components
Recognizable Entity Score 0 Practical Impact Score 44 Novelty Interest Score 94 Consequence Score 18 Curiosity Score 16 Shareability Score 53

VQV surfaced this signal because it is recent, relevant to Open Source LLMs, connected to Scobleizer News Pulse (X).