Alibaba Cloud / Qwen Team model product
Qwen3.8 Max
8 Max is Qwen's flagship multimodal mixture-of-experts model for coding, research, professional workflows, and long-horizon agents.
Updated Aug 12, 2026. Default version: Qwen3.8 Max
LLMBoard score89.3Qwen3.8 Max
Coverage80%23 benchmark families
Context window1MTokens
Official input price$2Alibaba API
Qwen3.8 Max Specifications
Technical details for the model's default version.
- Version
- Qwen3.8 Max
- Released
- Aug 2, 2026
- Knowledge cutoff
- Unknown
- Parameters
- 2.4T
- Context window
- 1M
- Max output
- 131.1K
- Inputs
- image, pdf, text, video
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Qwen3.8 Max Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Qwen3.8 Max Benchmark Results
Benchmark scores for Qwen3.8 Max.
| AndroidWorld | 85.3% | 01 | 4 | 100.0% | C | |
| CoWorkBench | 74.8% | 01 | 3 | 100.0% | C | |
| ERQA | 77.8% | 01 | 23 | 100.0% | C | |
| HealthBench | 60.2% | 01 | 9 | 100.0% | C | |
| Humanity's Last Exam (with tools, text-only) | 56.2% | 01 | 2 | 100.0% | C | |
| IFBench | 82.8% | 01 | 29 | 100.0% | C | |
| LongBench v2 | 66.3% | 01 | 17 | 100.0% | C | |
| LVBench | 81.8% | 01 | 24 | 100.0% | C | |
| MobileWorld | 77.8% | 01 | 3 | 100.0% | C | |
| NL2Repo | 55.9% | 01 | 14 | 100.0% | C | |
| OSWorld-Verified | 86.1% | 01 | 23 | 100.0% | C | |
| PaperBench | 93.0% | 01 | 3 | 100.0% | C | |
| PerceptionBench | 63.5% | 01 | 2 | 100.0% | C | |
| RealWorldQA | 88.0% | 01 | 26 | 100.0% | C | |
| SkillsBench | 70.2% | 01 | 8 | 100.0% | C | |
| VideoMME w sub. | 90.4% | 01 | 10 | 100.0% | C | |
| Vision2Web | 69.0% | 01 | 2 | 100.0% | C | |
| MLS-Bench Lite | 41.0% | 02 | 3 | 50.0% | C | |
| ScreenSpot Pro | 84.5% | 03 | 25 | 91.7% | C | |
| SWE-Bench Pro | 67.7% | 04 | 45 | 93.2% | C | |
| Terminal-Bench 2.1 | 86.6% | 04 | 19 | 83.3% | C | |
| FrontierSWE | 73.5% | 05 | 15 | 71.4% | C | |
| MMMU-Pro | 82.3% | 05 | 66 | 93.8% | C | |
| DeepSWE 1.1 | 56.6% | 10 | 20 | 52.6% | C | |
| GPQA | 92.6% | 12 | 234 | 95.3% | C | |
| Humanity's Last Exam | 43.6% | 28 | 93 | 70.7% | C |
Qwen3.8 Max Arena Results
Preference and agent-evaluation results for the default version.
| vision style control | overall | 02 | 1301.3 | 5,344 | N/A | |
| webdev | overall | 03 | 1671.1 | 2,531 | N/A | |
| text | overall | 06 | 1491.6 | 6,404 | N/A | |
| vision | overall | 06 | 1314.6 | 5,344 | N/A | |
| text style control | overall | 09 | 1490.7 | 6,404 | N/A | |
| text factuality | overall | 10 | 1481.1 | 6,404 | N/A |
Qwen3.8 Max Runtime Performance
Provider-specific output speed and catalog latency for Qwen3.8 Max. Runtime does not affect the capability score.
No runtime data
Browse runtime rankingsNo provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Qwen3.8 Max Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $2 input, $6 output per 1M
- Official provider
- Alibaba
- Lowest third-party
- From $1.65 input, $4.95 output per 1M via Deep Infra
- Tracked offerings
- 17
| Alibaba Token Plan | qwen3.8-max | global | N/A | N/A | 1M | |
| Alibaba Token Plan (China) | qwen3.8-max | global | N/A | N/A | 1M | |
| Deep Infra | Qwen/Qwen3.8-Max | global | $1.65 | $4.95 | 256K | |
| Alibaba (China) | qwen3.8-max | global | $1.78 | $5.33 | 1M | |
| LLM Gateway | qwen3.8-max | global | $1.82 | $5.45 | 1M | |
| CrossModel | qwen/qwen3.8-max | global | $1.88 | $5.63 | 1M | |
| OpenCode Go | qwen3.8-max | global | $2 | $6 | 1M | |
| NanoGPT | qwen3.8-max | global | $2 | $6 | 991K | |
| EmpirioLabs AI | qwen3-8-max | global | $2 | $6 | 1M | |
| Ofox | bailian/qwen3.8-max | global | $2 | $6 | 1M | |
| Kilo Gateway | qwen/qwen3.8-max | global | $2 | $6 | 1M | |
| Merge Gateway | qwen/qwen3.8-max | global | $2 | $6 | 1M | |
| Charm Hyper | qwen3.8-max | global | $2 | $6 | 1M | |
| Alibaba | qwen3.8-max | global | $2 | $6 | 1M | |
| Vercel AI Gateway | alibaba/qwen3.8-max | global | $2 | $6 | 1M | |
| OpenRouter | qwen/qwen3.8-max | global | $2 | $6 | 1M | |
| Venice AI | qwen-3-8-max | global | $2.5 | $7.5 | 1M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Qwen3.8 Max Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| Qwen3.8 Max | 89.3 | 2.4T | 1M | 131.1K | No | Proprietary |
What is Qwen3.8 Max?
Key information about Qwen3.8 Max and its available data.
8 Max is Qwen's flagship multimodal mixture-of-experts model for coding, research, professional workflows, and long-horizon agents. 4 trillion total parameters with 95 billion active parameters, accepts text and image input, and supports a 1 million token context window through QwenCloud.
Data as of 2026-08-11.
Qwen3.8 Max vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Qwen3.8 Max
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
AC
Qwen3.7 Max
Alibaba Cloud / Qwen Team
AC
Qwen3.7 Plus
Alibaba Cloud / Qwen Team
AC
Qwen3.6 Plus
Alibaba Cloud / Qwen Team
AC
Qwen3.5 397B A17B
Alibaba Cloud / Qwen Team
AC
Qwen3.6 27B
Alibaba Cloud / Qwen Team
AC
Qwen3.5 122B A10B
Alibaba Cloud / Qwen Team
FAQ
Common questions about Qwen3.8 Max.
When was Qwen3.8 Max released?
Qwen3.8 Max's default version was released on Aug 2, 2026.
How much does Qwen3.8 Max cost?
Qwen3.8 Max's official API price is $2 per million input tokens and $6 per million output tokens via Alibaba. The lowest tracked third-party offer starts at $1.65 input and $4.95 output via Deep Infra.
Who created Qwen3.8 Max?
Qwen3.8 Max was created by Alibaba Cloud / Qwen Team.
What is the context window for Qwen3.8 Max?
The default version has a 1M token context window.
Is Qwen3.8 Max open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Qwen3.8 Max?
17 provider offerings are linked to the default version.
What models should I compare Qwen3.8 Max with?
Nearby ranked alternatives include GPT-5.6-Sol, Claude Mythos, Kimi K3.