Alibaba
160 models across 32 families · Latest: Qwen4-Max-Preview (2026-11)
AI research institute of Alibaba Group.
Alibaba's portfolio covers 153 active models across 32 current families, spanning coding, rag, and agents. Open a model detail page to compare provider routes and sourced benchmarks.
Covers 7 workload areas across 153 active tracked models; last verified 2026-11-11.
Use it for
- Teams evaluating coding, rag, and agents across this lab's releases
- Comparing model families before committing to a flagship
- Migration and pricing follow-ups across 153 tracked models
Do not use it for
- Choosing a hosting provider without opening a model page for price ladders
Active models
153
Current models from this lab, excluding deprecated ones
Active families
32
Current model families from this lab
Open catalog
127 open
103 open source / 24 open weights
Lowest output price
$0.030 /1M
Cheapest tracked output across active models, per 1M tokens
Latest dated release
2026-11-11
Qwen4-Max-Preview
Freshness
2026-11-11
Researched 3d ago
Information
Release cadence
- Dated releases
- 5
- Latest model
- Qwen4-Max-Preview
- Latest date
- 2026-11-11
Where this lab wins
- Coding: 54 tracked models with SWE-bench / HumanEval-style scores.
- RAG: 49 tracked models with ruler / needle retrieval benchmarks.
- Agentic: 37 tracked models with BFCL, tau-bench, and SWE-bench tool-use coverage.
- Long-context: 92 tracked models with context-token or InfiniteBench-class signal.
Flagship quality / price signal
- Flagship
- Qwen2.5-7B-Instruct
- Selection source
- Best sourced coding Q/$
- Coding grade
- A
- Benchmark / output price
- humaneval 68.4 · $0.030/1M
About
Alibaba DAMO Academy is the AI research institute of Alibaba Group, dedicated to developing cutting-edge AI technologies and applications across various domains. Their work encompasses natural language processing, computer vision, machine learning, and robotics, with the goal of advancing AI capabilities and driving innovation across Alibaba's ecosystem of products and services. One notable research breakthrough is their development of a multimodal AI model that can understand and generate content across text, images, and videos, enabling more intelligent and interactive user experiences.
Featured models
| Model | Released | Context | Input price ($/1M) | Output price ($/1M) | License | Openness |
|---|---|---|---|---|---|---|
| Qwen4-Max-Preview | 2026-11-11 | 1M | - | - | Proprietary | Proprietary |
| Qwen-Image-2.1-Turbo | 2026-10-09 | - | - | - | Qwen Research License | Open weights |
| Qwen-Image-2.1-Pro | 2026-10-02 | - | - | - | Proprietary | Proprietary |
Model families
Recent releases
- Qwen4-Max-Preview- 2026-11-11
- Qwen-Image-2.1-Turbo- 2026-10-09
- Qwen-Image-2.1-Pro- 2026-10-02
- Qwen-Image-2.1- 2026-09-20
- Qwen3.8-LiveTranslate- 2026-09-19
Top comparisons
Explore related pages
Last reviewed: 2026-11-11. Data sourced from public lab announcements and provider documentation.