MiMo-V2.6-Pro
Grok 4.7
- Context
- 262Ktokens
- Released
- 6mo ago
Qwen3.6 Max Preview is an early preview of Alibaba's next proprietary model in the Qwen series, succeeding Qwen3.6-Plus. Alibaba says it improves on that predecessor in agentic coding, claiming the top score on six major coding benchmarks, and reports stronger world knowledge and better instruction following. The maker notes the model is still under active development, with further improvements expected in later versions.
- Language model
- Reasoning
- Tools
How it compares
Level
FlagshipThe lab's most capable generally available line, priced for hard tasks.
Flagship models from other labs
Vs. your stack
Add the models you use to see how this one compares.
Sign inAdditional source evidence
- Source model
- alibaba/qwen3.6-max-preview
- Source family
- qwen
- Last updated
- 2026-04-20
- Knowledge cutoff
- 2025-04
- Model context
- 262K
- Model output limit
- 66K
- Open weights
- No
- File attachments
- No
- Reasoning
- Yes
- Tool calling
- Yes
- Temperature control
- Yes
Complete source record (JSON)
{
"id": "alibaba/qwen3.6-max-preview",
"name": "Qwen3.6 Max Preview",
"description": "Flagship Qwen model for complex reasoning, coding, and agentic workflows",
"family": "qwen",
"last_updated": "2026-04-20",
"knowledge": "2025-04",
"attachment": false,
"reasoning": true,
"tool_call": true,
"temperature": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 262144,
"output": 65536
},
"release_date": "2026-04-20",
"open_weights": false
}Frequently asked questions
Capabilities and specs
The context window of Qwen3.6 Max Preview is 262K tokens (262,144). Its maximum output is 66K tokens (65,536).
No, Qwen3.6 Max Preview accepts text only. It cannot read images, audio, video or PDFs.
Yes, Qwen3.6 Max Preview supports tool calling (function calling).
Yes, Qwen3.6 Max Preview supports a reasoning mode.
The knowledge cutoff of Qwen3.6 Max Preview is 2025-04.
Quality and benchmarks
On Arena Text, Qwen3.6 Max Preview has rank 62 with a score of 1460. On Arena WebDev, it has rank 75 with a score of 1482. Both ranks are Arena's own rank.
Yes, Qwen3.6 Max Preview is ahead of its predecessor Qwen3 Max Preview on Arena Text. It is behind on none and level on none.
The closest alternatives to Qwen3.6 Max Preview are ERNIE 5.1 by Baidu, GPT-5.5 Instant by OpenAI and MAI-Code-1-Flash by Microsoft. They are matched on model kind and release date, with no comparable price.
ERNIE 5.1Qwen3.6 Max Preview vs ERNIE 5.1GPT-5.5 InstantQwen3.6 Max Preview vs GPT-5.5 InstantMAI-Code-1-FlashQwen3.6 Max Preview vs MAI-Code-1-Flash
Qwen3.6 Max Preview itself is closed. The highest-ranked open-weight model on Arena Text is Kimi K3, at rank 19.
Versions and lineage
Both models have the same 262K token (262,144) context. Qwen3.6 27B costs $0.32 input and $3.20 output per 1M tokens, and was released April 22, 2026, while Qwen3.6 Max Preview was released April 20, 2026.
Licensing and access
No, Qwen3.6 Max Preview is closed. It is neither open source nor open weights.
Basics
Qwen3.6 Max Preview was made by Qwen and released on April 20, 2026.
Generated from Vertical Tabs data, updated November 12, 2026.
Discussion
No comments yet
Be the first to share your thoughts.