New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Qwen3.8 Max
About Qwen3.8 Max
Qwen3.8 Max is an AI model developed by Alibaba Qwen, in the multimodal category, released in 2026, made available as a proprietary (API-only) model with 2400B params. Alibaba's Qwen3.8-Max is the full general-availability release of the Qwen3.8-Max family, officially launched August 2, 2026. Built on the Qwen 3.5 architectural foundation, it scales to 2.4 trillion total parameters (95B active in a mixture-of-experts design) and supports a 1-million-token context window with native multimodality (text, images, video, documents). It is the first Qwen-Max-class model announced for open-weight release. Key advances include autonomous long-horizon coding (10+ day self-evolving harness), dynamic-workflow orchestration for real-world work, multimodal agents with visual feedback loops, and frontier-level performance on PaperBench (93.0), Terminal-Bench 2.1 (86.6), SWE-bench Pro (67.7), and GPQA-Diamond (92.6). Available via QwenCloud API and compatible with Codex, Qwen Code, OpenClaw, and Hermes agent harnesses.
On this page you'll find Qwen3.8 Max's full specifications, including a context window of 1M tokens, 4 benchmark results. Review provider pricing and benchmark scores below, or compare Qwen3.8 Max head-to-head with other multimodal models.
Community
No reports yet. Own this model on some hardware? Be the first to confirm it runs.
No comments yet. Start the discussion.
Qwen3.8 Max benchmark scores
Compare on the benchmark table →Scores on standardised evaluations. Higher is better, and the percentile shows where Qwen3.8 Max lands among every model we score on that benchmark.
Benchmarks only say so much. Have you run Qwen3.8 Max yourself? Tell everyone how it went — what it is good at, where it fell over, and on what hardware.
More models like Qwen3.8 Max
Frequently asked questions
What is Qwen3.8 Max? ▶
Alibaba's Qwen3.8-Max is the full general-availability release of the Qwen3.8-Max family, officially launched August 2, 2026. Built on the Qwen 3.5 architectural foundation, it scales to 2.4 trillion total parameters (95B active in a mixture-of-experts design) and supports a 1-million-token context window with native multimodality (text, images, video, documents). It is the first Qwen-Max-class model announced for open-weight release. Key advances include autonomous long-horizon coding (10+ day self-evolving harness), dynamic-workflow orchestration for real-world work, multimodal agents with visual feedback loops, and frontier-level performance on PaperBench (93.0), Terminal-Bench 2.1 (86.6), SWE-bench Pro (67.7), and GPQA-Diamond (92.6). Available via QwenCloud API and compatible with Codex, Qwen Code, OpenClaw, and Hermes agent harnesses. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.
Who created Qwen3.8 Max? ▶
Qwen3.8 Max was developed by Alibaba Qwen and released in 2026.
Is Qwen3.8 Max open source or proprietary? ▶
Qwen3.8 Max is a proprietary model. It is accessed through an API rather than by downloading the weights.
What is Qwen3.8 Max's context window? ▶
Qwen3.8 Max supports a context window of 1M tokens, and has 2400B params.
How much does Qwen3.8 Max cost? ▶
Pricing for Qwen3.8 Max depends on the provider. See the providers table on this page for the latest API rates.
How does Qwen3.8 Max perform on benchmarks? ▶
Qwen3.8 Max is benchmarked across 4 evaluations on GenAIList, including Terminal-Bench 2.1 (86.6%). See the full benchmark table below and compare it with other models.
Know a great video about Qwen3.8 Max?
Add YouTube Video
Add Tweet/Discussion
Report an Issue
Help us maintain accurate information in our database.
Suggest an Edit
Tell us what needs to be changed or corrected for