All AI models · LLM benchmarks · Methodology

Alibaba

Qwen3-VL 30B-A3B

Qwen vision-language model for visual reasoning, documents, and agent tasks

Model facts

Context window
262K tokens
Maximum output
33K tokens
Cheapest paid input
$0.108 per 1M tokens
Cheapest paid output
$0.431 per 1M tokens
Open weights
yes
Providers
5

Capabilities and modalities

reasoning, tool calling, structured output, attachments, temperature control, open weights, text, image.

Local hardware estimate

At 4K context: Q4_K_M 17.6 GiB, Q8_0 32.1 GiB, F16 59.2 GiB working memory.

Parameters
31.1 billion
Architecture
qwen3vlmoe
Model license
apache-2.0

Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.

Artificial Analysis benchmarks

Compare this model on the full LLM benchmark leaderboard.

ConfigurationAAICodingMathOutput tokens/s
Reasoning13.4not measured82.3not measured

Observed price history

  • 2026-08-24: $0.108 input / $0.431 output per 1M tokens

API providers

  • Alibaba (China) — model id qwen3-vl-30b-a3b; $0.108 input / $0.431 output per 1M tokens; provider documentation
  • NEAR AI Cloud — model id Qwen/Qwen3-VL-30B-A3B-Instruct; $0.15 input / $0.55 output per 1M tokens; provider documentation
  • DevPass (LLM Gateway) — model id qwen3-vl-30b-a3b-instruct; $0.15 input / $0.6 output per 1M tokens; provider documentation
  • OpenRouter — model id qwen/qwen3-vl-30b-a3b-instruct; $0.15 input / $0.6 output per 1M tokens; provider documentation
  • Alibaba — model id qwen3-vl-30b-a3b; $0.2 input / $0.8 output per 1M tokens; provider documentation