Qwen3.8
Key facts · verified Nov 5, 2026
Qwen3.8 is an Alibaba model released on 3 Aug 2026. Its API price is $2 per million input tokens and $6 per million output tokens (Qwen Cloud). Its context window is 1M tokens (hosted Qwen3.8-Max); open weights 262k native, extensible to ~1.01M. It ranks #24 on LMArena's text leaderboard (Elo 1482, read 5 Nov 2026). Source: qwencloud.com.
Alibaba’s Qwen3.8-Max, previewed 19 July and launched 3 August 2026 as the largest model in the Qwen family: a mixture-of-experts design with 2.4 trillion total parameters and 95 billion active per query, across 512 experts. Alibaba published the weights on 12 August as Qwen/Qwen3.8-2.4T-A95B under a licence named "qwen3.8-max" - not Apache 2.0 - making it the largest downloadable model on our roster. It is text-only despite widespread reporting to the contrary: the multimodal member of the family is the separate Qwen3.8-27B. Context is 262,144 tokens natively, extensible to about 1.01M, and thinking mode is always on rather than optional.
$2 / $6 per million tokens
1M tokens (hosted Qwen3.8-Max); open weights 262k native, extensible to ~1.01M
Key Features
Full Specifications
$2 / $6 per million tokens | |
|---|---|
| Identity | |
| Developer | Alibaba |
| Released | Aug 2026 |
| Status | GA |
| Licence | Open weights under a licence named "qwen3.8-max" - NOT Apache 2.0 (the Qwen3.8-27B sibling is Apache 2.0, which is a common source of confusion) |
| Self-hostable | Yes - Qwen/Qwen3.8-2.4T-A95B and an FP8 checkpoint published 12 Aug 2026 |
| Cost | |
| Blended $/1M tokensinput × 0.75 + output × 0.25 | $3.000 / 1M tokens |
| Input price | $2.000 / 1M tokens |
| Output price | $6.000 / 1M tokens |
| Cached input | $0.25 / 1M tokens (implicit cache); explicit cache $2.50 to create, $0.17 to read |
| Batch discount | Batches supported on Qwen Cloud; discount not listed on the model page |
| Free tier | Not documented |
| Capacity | |
| Context window | 1M tokens (hosted Qwen3.8-Max); open weights 262k native, extensible to ~1.01M |
| Max output | 131K tokens (hosted) |
| Long-context surcharge | Not documented |
| Capability | |
| Vision in | Yes - the hosted Qwen3.8-Max takes text, image and video input. The open-weight Qwen3.8-2.4T-A95B is text-only ("Multimodal inputs are not supported"). |
| Audio in | No |
| Function calling | Yes |
| Structured output | Yes |
| Extended reasoning | Yes - the hosted Qwen3.8-Max supports thinking and non-thinking; the open weights require thinking |
| Web search | Yes - built-in web_search tool (hosted) |
| Code execution | Not documented |
| Access | |
| API | Yes |
| Chat app | Yes |
| Cloud marketplaces | Alibaba Cloud Model Studio |
| Fine-tuning | Yes - listed on Qwen Cloud; open weights also permit it |
| Measured quality | |
| LMArena Elo | 1482 (checked Nov 2026) |
| LMArena rank | Rank 24 |
| Elo per dollarLMArena Elo ÷ blended $/1M tokens | 494 |