NEWMEN
← back to all models

Qwen: Qwen3 VL 30B A3B Instruct

qwen/qwen3-vl-30b-a3b-instruct

Alibaba (Qwen)open-weight262K context6 providersIntelligence 58.0· est.

Cheapest provider

$0.13 / 1M

Alibaba

Fastest provider (p95)

84 tok/s

Alibaba

Intelligence (estimated)

58.0

Family + generation + popularity + price tier

Per-provider performance

Latency / throughput / uptime / price measured across providers over the last 30 minutes of live traffic. This is what proves “sourced cheapest” — Atlas mode draws on these per call to serve the cheapest path that holds quality.

ProviderQuantInput $/1MOutput $/1MLatency p50 / p95Throughput p50 / p95Uptime 30mSuccess
Alibabaundisclosed$0.1300$0.5200677ms / 1739ms46 / 84 tok/s100.00%100.0%
AtlasCloudq8· fp8$0.1500$0.6000985ms / 2474ms6 / 38 tok/s95.62%92.9%
DeepInfraq8· fp8$0.1500$0.6000330ms / 1219ms38 / 51 tok/s95.86%94.9%
Novitafull· bf16$0.2000$0.70002068ms / 5376ms23 / 40 tok/s88.79%88.8%
Phalaundisclosed$0.2000$0.7000863ms / 1655ms39 / 72 tok/s99.85%99.9%
SiliconFlowq8· fp8$0.2900$1.00004059ms / 6431ms9 / 14 tok/s95.58%94.8%

“—” means live telemetry hasn’t accumulated enough recent traffic for that endpoint. “undisclosed” means the provider serves the model but doesn’t expose the quantization label (typically running fp8 / int8 internally).

Intelligence estimate

No public benchmark numbers indexed for this model yet, so the leaderboard score is derived from the catalogue data we sync: model family standing, generation, and live usage rank — applied identically to open- and closed-weight models. The heuristic ceiling sits below confirmed-benchmark frontier models so curated rankings stay clearly on top.

Want a curated score? File a benchmark report and we’ll add it to the next sync.

How Atlas mode sources Qwen: Qwen3 VL 30B A3B Instruct

  • Strict mode — pin Qwen: Qwen3 VL 30B A3B Instruct exactly and we pass it straight through, sourced from the cheapest provider above. The same model, no substitutions — currently Alibaba at $0.13/1M.
  • Atlas mode — the default. Each call is auto-optimized for the cheapest path that holds quality, at least 5% off going direct from call one and climbing as it ramps. You always see which model served the call and exactly what you saved — thumbs-down anything you don’t like for a full refund.