GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Musci-research
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Musci-research
Model Tree
Qwen/Qwen3-ASR-1.7B-hf
Input Modalities
Output Modalities
Supported Functionality
The upstream Qwen3-ASR model is released under the Apache License 2.0. Please retain the upstream attribution and comply with the Apache-2.0 terms when using or redistributing this derivative model. Musci-ASR-2B is an unofficial fine-tuned model and is not an official Qwen release.
This checkpoint is fine-tuned from Qwen3-ASR-1.7B-HF with GRPO on an internal mixture built from the public Open ASR Leaderboard training prompts and curated hard-case speech examples. The released checkpoint is an unofficial research fine-tune; the benchmark test sets are not used for optimization.