✳ the wire · intel

Qwen 3.8 Max preview is live — 2.4T params, open weights promised

Qwen 3.8 Maxconfirmedby ArtificialWatch
Alibaba Qwen announcement art
Source imagery · verified against a primary source

Alibaba launched the Qwen3.8-Max preview Jul 18 (Jul 19 China time), alongside its new Token Plan individual subscription. The headline number: 2.4 trillion total parameters — a big jump from Qwen3.7-Max, which shipped less than two months ago, and the same scale bracket as Kimi K3's 2.8T.

The team's own positioning is aggressive: "one of the most powerful models available today… second only to Fable 5." That is self-reported — no independent board has scored it yet. We'll fold it into the Bench Index the moment LMArena, LiveBench or Artificial Analysis publish measured runs.

The rest of this analysis — 1 more paragraph — plus the source, is on the paid plans.

Every wire item is source-verified before it posts and graded: confirmed · strong · reported · rumor. Subscribers see the primary source on each one, the reasoning behind its grade, and the full analysis — plus the launch alerts themselves, which fire the minute a model answers rather than whenever the feed catches up.

See the plans →

sweeping every 60 seconds

Know the minute it drops — not the minute we write it up.

Claude Opus 5 went live at 16:51 UTC. The alert was in subscribers' inboxes at 16:52.

  • Free forever
  • No card
  • Unsubscribe in one click

← back to the wire