Releases·Model Release
Alibaba announces Qwen 4 family at Apsara, says open-weights release is 'very soon'
Alibaba announced the Qwen 4 family at its Apsara conference, with a Medium writeup separating confirmed details from rumor. The open-weights tier is described as coming "very soon," and the architecture reportedly already runs on consumer gaming GPUs.
0 engagement1 sourceFri, Oct 2, 2026, 09:00 PM
AI Overview
- Alibaba announced the Qwen 4 model family at its Apsara conference, per a Medium post dated 2026-10-02
- The post says the open-weights tier of Qwen 4 is coming 'very soon'
Entities
Related
Releases·Model Releasefeatured
Alibaba Announces Qwen 4 at Apsara Conference
Alibaba officially announced Qwen 4 at its Apsara Conference. The announcement is newsworthy to practitioners tracking the Qwen model family's next generation, though the posts contain no further technical specifics.
Alibaba officially announced Qwen 4 at its Apsara Conference. The announcement is newsworthy to practitioners tracking the Qwen model family's next generation, though the posts contain no further technical specifics.
2.4k engagement1 sourceredditTue, Sep 22, 2026, 02:45 AM
Social·Community
Qwen 4 reportedly planned for end of October, per Alibaba 0-day partner
A Reddit user says a self-described 0-day partner of Alibaba casually indicated that Qwen 4 is planned for release at the end of October, sooner than the user expected given the gap between Qwen 3.6 and 3.8. The claim is unverified rumor, acknowledged by the poster as 'in bro we trust.' Practitioners may treat it as a soft signal, not confirmation.
A Reddit user says a self-described 0-day partner of Alibaba casually indicated that Qwen 4 is planned for release at the end of October, sooner than the user expected given the gap between Qwen 3.6 and 3.8. The claim is unverified rumor, acknowledged by the poster as 'in bro we trust.' Practitioners may treat it as a soft signal, not confirmation.
766 engagement1 sourceredditTue, Oct 6, 2026, 12:48 PM
Releases·Model Releasefeatured
Qwen releases Qwen3.8-Flash-Next, previewing Qwen 4 architecture
Qwen released Qwen3.8-Flash-Next, an open-weight hybrid model that previews the Qwen 4 architecture. The 125B MoE model reportedly outperforms DeepSeek V4 Flash at half the parameters and is stronger than Opus 4.6. Its use of n-gram tables could enable running 1T+ parameter models on a single server with modest GPUs and system RAM.
Qwen released Qwen3.8-Flash-Next, an open-weight hybrid model that previews the Qwen 4 architecture. The 125B MoE model reportedly outperforms DeepSeek V4 Flash at half the parameters and is stronger than Opus 4.6. Its use of n-gram tables could enable running 1T+ parameter models on a single server with modest GPUs and system RAM.
41009.0k engagement14 sourcesreddit, x, rss +1Wed, Aug 26, 2026, 01:36 PM
Developments·Hardwarefeatured
GPUs could explode to multiple TB with new storage-inspired memory tech
A new storage-inspired memory technology could enable GPUs to reach multiple terabytes of memory, according to a report from The Register. The approach leverages storage-class memory concepts to dramatically increase GPU memory capacity, potentially transforming AI workloads that are currently limited by VRAM sizes.
A new storage-inspired memory technology could enable GPUs to reach multiple terabytes of memory, according to a report from The Register. The approach leverages storage-class memory concepts to dramatically increase GPU memory capacity, potentially transforming AI workloads that are currently limited by VRAM sizes.
44 engagement1 sourcehackernewsSun, Aug 2, 2026, 12:06 PM
Social·Community
Community speculates on Qwen 4 architecture
A Reddit user speculates about the architecture of the upcoming Qwen 4 model, proposing several possible designs including embedding-offloaded linear layers with 51b n-grams, sparse full-attention with dense routing, and a multi-head latent attention hybrid. The post reflects community interest and speculation ahead of an official release.
A Reddit user speculates about the architecture of the upcoming Qwen 4 model, proposing several possible designs including embedding-offloaded linear layers with 51b n-grams, sparse full-attention with dense routing, and a multi-head latent attention hybrid. The post reflects community interest and speculation ahead of an official release.
1.2k engagement2 sourcesredditTue, Aug 25, 2026, 01:15 PM