Qwen3.8 2.4T-A95B
Qwen (Alibaba) · #7 in our ranking · Best for multilingual
Qwen3.8 2.4T-A95B is an open-weight mixture-of-experts model from Qwen (Alibaba), released Aug 2026, with 2.4T parameters (95B active per token), a 256K-token context window, and the Qwen3.8-Max License.
Qwen3.8 2.4T-A95B is the open-weight model behind Qwen3.8-Max. Alibaba reports large gains in coding, professional work, and long-horizon agent tasks over Qwen3.5 and 3.6, with tunable reasoning effort.
Qwen3.8 2.4T-A95B specs
- Provider
- Qwen (Alibaba)
- Architecture
- Mixture-of-experts
- Parameters
- 2.4T total · 95B active
- Context
- 256K tokens | 197K words
- License
- Qwen3.8-Max License
- Input
- Text
- Released
- Aug 2026
- Weights
- Hugging Face
Sourced from the model card on Hugging Face. Last checked 2026-10-01.
Highlights
- 2.4T total, 95B active
- Qwen-Max class, open weights
- Tunable reasoning effort
Where it fits for payment teams
- AML and compliance investigations
#3 pick. Qwen-Max-class quality with broad multilingual coverage for cross-border alerts.
- Payments customer support
#3 pick. Broad multilingual coverage for global merchant bases.
Harnesses in Runtime
Run Qwen3.8 2.4T-A95B through any of these harnesses on Runtime.
How to run Qwen3.8 2.4T-A95B
Download the weights
Pull Qwen/Qwen3.8-2.4T-A95B from Hugging Face. Check the Qwen3.8-Max License before you deploy.
Serve it in your cloud
Run it with vLLM or SGLang on your own GPUs, or use a managed provider that hosts the model.
Put it to work in Runtime
Point an agent or a single skill at the model through OpenCode, then compare it to your current model with evals.
Qwen3.8 2.4T-A95B FAQ
What is the context window of Qwen3.8 2.4T-A95B?
+
Qwen3.8 2.4T-A95B supports a 256K-token context window (262,144 tokens), roughly 197K words of English text.
How many parameters does Qwen3.8 2.4T-A95B have?
+
Qwen3.8 2.4T-A95B is a mixture-of-experts model with 2.4T total parameters, of which 95B are active per token.
Is Qwen3.8 2.4T-A95B free for commercial use?
+
Qwen3.8 2.4T-A95B is released under the Qwen3.8-Max License, a custom license. Commercial use may come with conditions, so read the license on Hugging Face before you deploy or fine-tune it.
What can Qwen3.8 2.4T-A95B take as input?
+
Qwen3.8 2.4T-A95B accepts text input and generates text.
Is Qwen3.8 2.4T-A95B good for AML and compliance investigations?
+
Yes. Qwen3.8 2.4T-A95B is our #3 pick for AML and compliance investigations. Qwen-Max-class quality with broad multilingual coverage for cross-border alerts.
Is Qwen3.8 2.4T-A95B good for payments customer support?
+
Yes. Qwen3.8 2.4T-A95B is our #3 pick for payments customer support. Broad multilingual coverage for global merchant bases.
Can I run Qwen3.8 2.4T-A95B in my own cloud with Runtime?
+
Yes. Download the weights from Hugging Face (Qwen/Qwen3.8-2.4T-A95B) and serve them with vLLM or SGLang in your cloud, or use a managed provider that hosts the model. Runtime agents can then use Qwen3.8 2.4T-A95B through a harness like OpenCode, with your data staying in your environment.