Qwen 3.6 Max Preview
Qwen 3.6 Max Preview is Alibaba's early-access release of the Qwen 3.6 Max line, a preview build that focuses on coding-agent execution, front-end development quality, and long-tail knowledge over a context window of 240K tokens.
import { streamText } from 'ai'
const result = streamText({ model: 'alibaba/qwen-3.6-max-preview', prompt: 'Why is the sky blue?'})Playground
Try out Qwen 3.6 Max Preview by Alibaba. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Providers
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
More models by Alibaba
| Model |
|---|
| 1M | 1.1s | 118tps | $2.50/M | $7.50/M | Read:$0.25/M Write:$3.13/M | — | — |   | 05/21/2026 | ||||
| 1M | 1.6s | 109tps | $0.50/M | $3.00/M | Read: $0.1/M Write: $0.63/M | — | — |  | 04/02/2026 | ||||
| 1M | 1.2s | 167tps | $0.10/M | $0.40/M | Read:$0.0/M Write:$0.13/M | — | — | | 02/24/2026 | ||||
| 1M | 1.4s | 110tps | $0.40/M | $2.40/M | Read: $0.04/M Write: $0.5/M | — | — | | 02/16/2026 | ||||
| 256K | 1.2s | 32tps | $0.50/M | $1.20/M | — | — |   | 07/22/2025 | |||||
| 262K | 0.1s | 87tps | $0.07/M | $0.46/M | Read:$0.6/M Write:— | — | — |   +1 | 04/01/2025 |
About Qwen 3.6 Max Preview
Qwen 3.6 Max Preview is the preview entry in the Qwen 3.6 Max line, succeeding the previously released Qwen3-Max and complementing the Plus-tier Qwen3.6-Plus. Compared with those models, Qwen 3.6 Max Preview targets a different center of gravity: coding agents that need to plan and act over long horizons, front-end development work that demands accurate component output, and broader long-tail knowledge retention.
The model is served as a closed-weight API release with a context window of 240K tokens and a maximum output of 64K tokens, which gives it room for full repository context, multi-file diffs, and extended agent traces without segmentation. Tagged capabilities include reasoning, tool use, implicit caching, file input, and vision, so it can ingest images and document files alongside text within the same request.
The preview designation signals that Alibaba is iterating on Qwen 3.6 Max Preview ahead of its production release. Teams that integrate against the preview can validate prompt templates, evaluate output formatting for front-end scaffolding, and benchmark coding-agent loops against the new capability profile while AI Gateway handles provider routing and authentication. Because AI Gateway abstracts the underlying endpoint, migrating to the GA model identifier is a single configuration change.
What To Consider When Choosing a Provider
- Configuration: Preview models can shift in capability or rate-limit behavior before general availability. Validate prompts and tool schemas against the preview, but plan a regression pass before migrating production traffic to the GA identifier.
- Zero Data Retention: AI Gateway does not currently support Zero Data Retention for this model. See the documentation for models that support ZDR.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
When to Use Qwen 3.6 Max Preview
Best For
- Long-Horizon Coding Agents: Pipelines that plan and act across many tool-call turns with a context window of 240K tokens
- Front-End Development Work: Tasks where the model produces accurate component code, layout, and styling output
- Pre-GA Evaluation: Validating prompts and tool schemas against a near-final Max-tier model before production rollout
- Multimodal Coding Workflows: Combining screenshots, design references, and text in a single request via file input and vision
- Repository And Document Ingestion: Long-context tasks that load full codebases or document plus image inputs in one call
Consider Alternatives When
- GA Stability Required: Wait for the Qwen 3.6 Max general availability release when production guarantees are needed
- Smaller Tier Sufficient: Qwen3.6-Plus is a closer fit when the Plus line already meets your quality bar at lower cost
- Visible Reasoning Traces: Qwen3-Max-Thinking is a stronger match when chain-of-thought transparency drives evaluation
- Text-Only Pipelines: A dedicated text model offers lower cost per token when vision and file input go unused
Conclusion
Qwen 3.6 Max Preview extends the Qwen Max line with coding-agent and front-end focus, available through AI Gateway ahead of general availability. Routing through AI Gateway means migrating from the preview identifier to the GA model is a configuration change rather than an integration rewrite.
Frequently Asked Questions
What is the relationship between Qwen 3.6 Max Preview and Qwen3-Max?
Qwen 3.6 Max Preview is the preview release of the next generation in the Max tier, succeeding Qwen3-Max. Alibaba describes it as adding stronger vibe-coding ability, more efficient coding-agent execution, and improved front-end development skills, alongside upgraded long-tail knowledge retention.
What is the context window for Qwen 3.6 Max Preview?
The context window is 240K tokens and the maximum output is 64K tokens, which suits long-horizon agent traces, full-repository inputs, and multimodal requests that combine text with images or files.
Can Qwen 3.6 Max Preview accept images or files alongside text?
Yes. Qwen 3.6 Max Preview is tagged for vision and file input, so it can process images, documents, and text together within a single request through AI Gateway.
How do I integrate Qwen 3.6 Max Preview through AI Gateway?
Authenticate with an AI Gateway API key or OIDC token and reference `
alibaba/qwen-3.6-max-preview` as the model. You can call Qwen 3.6 Max Preview through AI SDK, Chat Completions API, Responses API, Messages API, or other API formats, from TypeScript or Python.Will prompts built against the preview work with the GA model?
In most cases yes, since the preview and GA share architecture and capability targets. Run regression tests before switching identifiers, as minor behavioral changes can occur between a preview build and its general availability release.
Does Qwen 3.6 Max Preview support zero data retention?
Zero Data Retention is not currently available for this model. Zero Data Retention is offered on a per-provider basis. See https://vercel.com/docs/ai-gateway/capabilities/zdr for details.
Where can I see live latency and cost data for Qwen 3.6 Max Preview?
This page shows live throughput, time-to-first-token, and pricing metrics for Qwen 3.6 Max Preview measured across real AI Gateway traffic.