← Speech-to-Text 
Alibaba
Qwen3-ASR
Measured 2026-07-03
* Conditional
Asia-hosted endpoint — p50 carries the network round-trip to Singapore.
WER · batch
2.8%
WER · stream
4.0%
WER · conversational
12.2%
The live socket runs qwen3-asr-flash-realtime. UNUSABLE FOR A CALL: median 311s of wall clock per 7s clip on this corpus, up to 1123s -- accuracy is competitive, responsiveness is not. n=97.
Finalize
424ms
p50–p90 · us-east4 · n=30
Time to first token
6.60s
p50–p90 · us-east4
Cost / min
$0.0054
qwen3-asr-flash-realtime, Singapore. The file/sync SKU is $0.0021.
Share this result
Embed the live badge
Markdown
[](https://benchmarks.speko.ai/stt/alibaba-qwen3-asr)
HTML
<a href="https://benchmarks.speko.ai/stt/alibaba-qwen3-asr"><img src="https://benchmarks.speko.ai/badge/stt/alibaba-qwen3-asr.svg" alt="Speko stt rank"></a>
URL
https://benchmarks.speko.ai/stt/alibaba-qwen3-asr