AIBRIEF
Web deploy platform Vercel and AI coding tool Command Code published day one pricing: $2 per million input tokens, $6 per million output, with 1M token context and image input support.
Alibaba released its new flagship foundation model, Qwen3.8, on Monday, and within hours more than half a dozen Western developer platforms had added it. The integrations are the launch-day signal: OpenRouter, Vercel, OpenCode, Command Code, Novita, and Hermes Agent all carry Qwen3.8 today, per leiphone's coverage of Alibaba's announcement and the platforms' own model pages.
Qwen3.8 is a Mixture-of-Experts (MoE) model: a design that activates only a fraction of its parameters per query, keeping inference cheap at scale. Alibaba says it has 2.4 trillion total parameters, 95 billion active per query, and a 1-million-token context window, with image-input support. Both Vercel's AI Gateway and Command Code published pricing on day one: $2 per million input tokens and $6 per million output tokens, 1M context, 128K max output.
Alibaba also reported leaderboard placements (TextArena second, CodeArena fourth, Vision Arena second) via leiphone. Those are Alibaba-via-media claims, not independent benchmark results, and "second to Claude" is contestable. The leaderboards themselves are where to check.
Open-source releases of Qwen3.8-Max and Qwen3.8-27B are planned for the following week. Same-day confirmations from OpenCode on X and Hermes Agent's creator anchor the integration list.