Qwen3 VL 30B A3B Instruct
by Prime Inference · family qwen · listed Oct 2025TVW
Open vision-language MoE model for image and video understanding, documents, and visual agents
Current rates USD per 1M tokens
Input$0.20
Output$0.70
Cache read—
Cache write—
Against the market
On the record
- Model id
qwen/qwen3-vl-30b-a3b-instruct- Context window
- 262K tokens
- Max output
- 16K tokens
- Input modalities
- text, image
- Output modalities
- text
- Reasoning
- no
- Tool calls
- yes
- Open weights
- yes
- Knowledge cutoff
- —
- Released
- 2025-10-04
- Last updated
- 2025-10-04
- Provider docs
- docs.primeintellect.ai/inference/overview
Back of the envelope
≈
Also listed at
- Merge Gateway$0.108 in · $0.43 out
- Deep Infra$0.15 in · $0.60 out
- NEAR AI Cloud$0.15 in · $0.55 out
- LLM Gateway$0.15 in · $0.60 out
- Kilo Gateway$0.15 in · $0.60 out
- OpenRouter$0.15 in · $0.60 out
- Eden AI$0.15 in · $0.60 out
- DevPass (LLM Gateway)$0.15 in · $0.60 out
- Novita AI$0.20 in · $0.70 out
- Hugging Face$0.20 in · $0.70 out
- SiliconFlow$0.29 in · $1.00 out
- SiliconFlow (China)$0.29 in · $1.00 out