Currently listed through these providers:
Model details
Qwen3.6 Max Preview
Qwen3.6 Max Preview represents Alibaba's most ambitious language model effort to date, built on a sparse mixture-of-experts architecture with approximately one trillion total parameters. The design philosophy centers on agentic coding and structured tool use, making it particularly suited for software engineering tasks that demand multi-step reasoning and external function integration. The model features a distinctive thinking mode that preserves reasoning traces across extended conversations, enabling developers to trace and audit the model's problem-solving path in complex, multi-turn agent workflows. This architecture choice reflects a deliberate balance between raw capability and the practical demands of deploying AI in developer tooling and automated coding environments.
A notable shift accompanies this release: for the first time in Qwen's history, the flagship model ships exclusively under closed weights, signaling Alibaba's strategic pivot toward proprietary deployment rather than open-source community contributions. Despite the closed approach, the model maintains broad accessibility through OpenAI and Anthropic API-compatible endpoints, lowering integration friction for teams already embedded in those ecosystems. Benchmark results published by independent reviewers place the model atop six major coding and reasoning evaluations, with the Artificial Analysis Intelligence Index rating it significantly above the median for reasoning models in its pricing tier. This combination of benchmark leadership, extensive context support, and thinking-mode preservation positions Qwen3.6 Max Preview as a strong fit for teams prioritizing agentic workflows and complex, long-horizon task completion.
Quick Info
Powered by- Provider
- Kilo Gateway
- Model key
- qwen/qwen3.6-max-preview
- Release date
- Apr 20, 2026
- Last updated
- Apr 20, 2026
- Knowledge cutoff
- 2025-04
- Input modalities
- Output modalities
- Capabilities
Cost
- Input token cost
- $1.027
- Output token cost
- $6.162
Limits
- Output tokens
- 65,536 tokens
- Context window
- 262,144 tokens