VercelEngineering PostOfficial Source

Qwen 3.8 Max now available on Vercel AI Gateway

Vercel AI Gateway now exposes Qwen 3.8 Max to coding agents, adding one model endpoint for long-context text and vision work with gateway routing, budgets, and usage tracking.

Vercel · Aug 2, 2026

Open Source Open Markdown Open JSON

Source Summary

Vercel AI Gateway now serves **Qwen 3.8 Max** under alibaba/qwen3.8-max. The model combines text and vision-language work, has **2.4 trillion parameters**, and supports up to **1 million tokens** of context.

Practical Implication

Builders can connect Claude Code, Codex, OpenCode, or Pi through the gateway setup command, then select the model for coding, screenshot-to-page, captioning, or image-grounded tasks. Gateway controls include usage and cost tracking, retries, failover, budgets, routing, and Zero Data Retention support.

Agent-Ready Context

Vercel AI Gateway now serves **Qwen 3.8 Max** under alibaba/qwen3.8-max. The model combines text and vision-language work, has **2.4 trillion parameters**, and supports up to **1 million tokens** of context.

Builders can connect Claude Code, Codex, OpenCode, or Pi through the gateway setup command, then select the model for coding, screenshot-to-page, captioning, or image-grounded tasks. Gateway controls include usage and cost tracking, retries, failover, budgets, routing, and Zero Data Retention support.

The announcement gives specifications and intended use cases, but no quality, latency, or agent benchmark results. The long context and parameter count do not establish whether it is a better default than models already in an agent stack.

Connected Context · Feed7 Judgment

This expands the gateway’s coding-model choice with one endpoint spanning text, vision, and very long context. It supports testing consolidated multimodal workflows through existing agent clients, but does not change the prior conclusion that routing should be workload-specific: parameter count and context length offer no comparative evidence on quality, latency, reliability, or cost.

Laguna S 2.1 is now available on AI GatewayLaguna’s paid route shares the 1M-context ceiling and coding-agent positioning, making it a relevant workload-level comparison rather than allowing context size alone to select Qwen.Kimi K3 and Kimi K3 Fast with ZDR and US-based providers now on AI GatewayKimi shows the operational dimensions absent from the Qwen announcement—speed tier, residency, retention, and price choices—clarifying what must be measured beyond model specifications.Claude Opus 5 now available on AI GatewayOpus adds reasoning controls, fast mode, and fallback considerations to the same gateway choice set, reinforcing that model selection depends on task and operating policy rather than headline scale.Alishahryar1/free-claude-codeThe local proxy offers similar client-side model portability without a managed gateway, contrasting Vercel’s bundled budgets, tracking, retries, failover, and retention controls with operator-owned maintenance and credential risk.

Context Map

toolscodingimage#model-selection#coding-agents#gateways

Uncertainty

The announcement gives specifications and intended use cases, but no quality, latency, or agent benchmark results. The long context and parameter count do not establish whether it is a better default than models already in an agent stack.