Models / Alibaba / Qwen 3.8 Max Prime
Alibaba · qwen3.8-max
Qwen 3.8 Max Prime
High-throughput edition of Qwen3.8 Max for coding, professional work, multimodal understanding, and long-running agent workflows
$4.00 in · $12.00 out
per 1M tokens · provider rate closest to the median. No first-party Alibaba price was listed. The reference price is Kilo Gateway's rate, the provider closest to the median.
Provider input prices run from $4.00 to $5.50 per 1M tokens. Output prices run from $12.00 to $16.50.
- Context
- 1,000,000
- Max output
- 131,072
- Cache read
- $0.50
- Cache write
- —
- Reasoning
- Yes
- Tool calls
- Yes
- Structured output
- Yes
- Vision
- Yes
- Open weights
- No
- License
- —
- Released
- Sep 23, 2026
- Updated in source
- Sep 23, 2026
- Knowledge
- —
- Providers
- 5
Providers
| Provider | Input | Output | Cache read | Cache write | Status |
|---|---|---|---|---|---|
| Kilo Gatewayreference | $4.00 | $12.00 | $0.50 | — | — |
| NanoGPT | $4.00 | $12.00 | $0.50 | — | — |
| OpenRouter | $4.00 | $12.00 | $0.50 | — | — |
| Vercel AI Gateway | $4.00 | $12.00 | $0.50 | $5.00 | — |
| AI/ML API | $5.50 | $16.50 | $0.69 | — | — |
Jobs this model qualifies for
- Coding agents — clears the bar
- Long-document Q&A — clears the bar
- Cheapest summarization — clears the bar