Qwen3.8 Flash
by IteraCompute · family qwen · listed Aug 2026RTV
Qwen vision-language model for visual reasoning, documents, and agent tasks
Current rates USD per 1M tokens
Input$0.11
Output$0.38
Cache read$0.016
Cache write—
Against the market
On the record
- Model id
qwen/qwen3.8-flash- Context window
- 1.0M tokens
- Max output
- 131K tokens
- Input modalities
- text, image
- Output modalities
- text
- Reasoning
- yes
- Tool calls
- yes
- Open weights
- no
- Knowledge cutoff
- —
- Released
- 2026-08-26
- Last updated
- 2026-08-26
- Provider docs
- iteracompute.com/docs.html
Back of the envelope
≈
Also listed at
- Alibaba Token Planfree in · free out
- Alibaba Token Plan (China)free in · free out
- SCNet Token Planfree in · free out
- Zenifra$0.109 in · $0.363 out
- Ofox$0.11 in · $0.39 out
- AIHubMix$0.113 in · $0.38 out
- Deep Infra$0.113 in · $0.382 out
- Runcrate$0.119 in · $0.401 out
- Alibaba (China)$0.119 in · $0.401 out
- Vancine$0.12 in · $0.38 out
- CrossModel$0.13 in · $0.43 out
- DevPass (LLM Gateway)$0.15 in · $0.47 out
- Alibaba$0.15 in · $0.47 out
- Merge Gateway$0.15 in · $0.47 out
- Boundless$0.15 in · $0.47 out