Qwen3.8 Max (0902)
alibaba/qwen3.8-max-0902- Alibaba
- Family · Qwen3 Max
- Text · Image · Video → Text
- Reasoning · adjustable
- Tools · JSON · Vision
Not on OSMEXO yet
Spec & list price
Context1M
Max output128K
Input / 1M$2.00
Output / 1M$6.003× input
Usage
Tokens / day24.6B▼ 14% 7d
Usage rank#67
Requests / day191.2K
Tool calls / req0.51
Per request · 30 days
Avg input54.1K3.6× median
Avg output1.7K2.5× median
Cache hit92%1.3× median
Reasoning share78%1.4× median
Tool-call failures2.5%1.3× median
Consumption
Peak 63.1B 12 OctLow 14.1B 25 OctRank today #67
News & trends
19 mentions this week ▼ 27%Qwen3 Max family · videos, posts, articles · 30 days
Trending · 14 days · Qwen3 Max family
- 3 AI Models FREE Until Nov 20! DeepSeek V4.1 Flash, GLM 5.3 Flash & Qwen 3.8 Max5.2K views
- 10 Million FREE Tokens for AI Coding 🤯 | GLM-5.3, DeepSeek V4.1, Qwen 3.8 Max & MiniMax M31.3K views
News · latest from Alibaba
- Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hardgithub.com · 31 Oct
- Qwen3.8-Flash Now Available in the EU on Alibaba Cloudtwitter.com · 31 Oct
- Alibaba's Qwen3.8-Flash-Next model tried to upload a file to Beijing in a testnotebookcheck.net · 27 Oct
- Running Qwen3.8-Flash-Next NVFP4 at 184 tok/s on a single RTX 6000 Promeshapplied.com · 26 Oct
Benchmarks
Benchmark data: Artificial AnalysisOutput speed#288/33434 tok/s1.9× slower than median
First token#2933.04 s3.0× longer than median
Family progress
Intelligence Index
- Family releases45.4
Technical details
- Reasoning
- on · xhigh
- Tokenizer
- Qwen
- Moderation
- none
Reasoning effort
- xhigh · default
- high
- medium
- low
- minimal
Request parameters 17
- tools
- tool_choice
- response_format
- structured_outputs
- reasoning
- reasoning_effort
- include_reasoning
- temperature
- top_p
- top_k
+7 more
- max_tokens
- stop
- seed
- frequency_penalty
- logprobs
- presence_penalty
- top_logprobs
When to use
Best for
- Hardest chat and reasoning tasks in the family
- Image and video understanding
- Long document analysis
- Tool-calling agents with structured output
Think twice for
- Self-hosting (closed weights)
- High-volume simple chat (flash options are cheaper)
- Audio input
Same family
Qwen3 Max- 21.3
- 28.4
- 15.6
- 45.4
- 29.5
Want Qwen3.8 Max (0902) on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.