Qwen3.6 Max Preview by Alibaba is an early preview proprietary model with improved agentic coding, stronger world knowledge, preserve_thinking API, and active iteration.
Capabilities, design details, and architectural traits
Qwen3.6 Max Preview is officially described as an early preview of Alibaba's next proprietary model - explicitly labeled as still under active development with further improvements expected. It sits above Qwen3.6-Plus in the 3.6 family and is text-only (no image or video input), unlike the multimodal Qwen3.6-Plus it succeeds. It is available via Alibaba Cloud Model Studio as qwen3.6-max-preview and is not open-weight.
| Trait | Detail |
|---|---|
| Agentic coding gains | Officially documented improvements across SkillsBench, SciCode, NL2Repo, and Terminal-Bench 2.0; text-only model focused on coding agent execution |
| Stronger world knowledge and instruction following | Documented gains in SuperGPQA, QwenChineseBench, and ToolcallFormatIFBench over Qwen3.6-Plus |
preserve_thinking feature | Carries <think> reasoning blocks across all preceding turns; explicitly recommended for agentic tasks |
| OpenAI and Anthropic API compatibility | Supported across Beijing, Singapore, and US-Virginia regional endpoints |
The "preview" status is central to this model's documented identity - Alibaba explicitly states it is iterating on the model and welcomes community feedback before a stable release.
Independent evaluations · Artificial Analysis
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | Alibaba | Anthropic | Anthropic |
| Release Date | April 20, 2026 | July 24, 2026 | June 9, 2026 |
| Knowledge Cutoff | - | May 2026 | - |
| Context & Limits | |||
| Context Window | 262K | 1M Best Context Window | 1M Best Context Window |
| Pricing (per 1M tokens) | |||
| Input Pricing | $1.30 Best Input Pricing | $5 | $10 |
| Output Pricing | $7.80 Best Output Pricing | $25 | $50 |
| Modalities | |||
| Inputs | text | textimage | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 41.1 | 63.1 Best Intelligence Index | 62.1 |
| Coding Index | - | 78.0 Best Coding Index | 76.5 |
| Agentic Index | - | 59.2 Best Agentic Index | 56.6 |
Graduate-level reasoning and expert Q&A evaluation.
Extremely difficult logical reasoning and knowledge.
Logical reasoning over long context windows.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
