Alibaba · audio
Qwen3 ASR 0.6B
Qwen3 ASR 0.6B is a downloadable speech-recognition checkpoint for audio transcription and language identification. The family documents 30 languages and 22 Chinese dialects, with streaming and offline inference paths.
Downloadable checkpoint · apache-2.0
Official sources reviewed 2026-10-02. This page documents access and specifications; it does not report a model benchmark.
Task and deployment contract
Downloadable checkpoint; run the task-specific inference code linked below.
The 0.6B checkpoint is one of the two released Qwen3-ASR sizes. Use the repository’s current task-specific runtime and audio preprocessing.
The family supports language identification and transcription. The language and dialect list is documented separately from general audio understanding models.
Word or segment alignment is a separate task. Qwen3-ForcedAligner-0.6B has its own model and language coverage; speaker diarization is not asserted for this checkpoint.
For long audio, follow the repository chunking and streaming guidance. No token context limit, real-time latency or paid endpoint is inferred from the model name.
Billing and limits
No current provider rate for this exact configuration is recorded here. Open weights still require compute and operations; an unlisted rate does not mean a hosted API is free.
Audio limits are task-specific; no text-token context window is inferred.
Fast Pivot — Access hundreds of AI models from a single API — competitive pricing, reliable uptime, enterprise-grade infrastructure.