GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
burakaydinofficial
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
burakaydinofficial
Model Tree
openai/whisper-medium
Input Modalities
Output Modalities
Supported Functionality
Evaluate this control and whisper-medium-mla-cv11 on CommonVoice-17 (scripts/validate.py in the
code repo); the per-language difference is the
conversion cost reported on the MLA card and in docs/results/.
| Lang | this control |
|---|---|
| en | 8.3 WER |
| de | 8.7 WER |
| es | 6.4 WER |
| fr | 13.4 WER |
| ru | 8.0 WER |
| tr | 13.1 WER |
| cy | 41.5 WER |
| ar | 23.9 WER |
| ka | 80.0 WER |
| th | 6.2 CER |
Encoder frozen during fine-tuning; 15,000 steps, warmup+cosine, fp16. Read-speech domain (CommonVoice + FLEURS-validated). "Compression cost" does not apply to this unconverted control.