GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ā
PleasedPenguin
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
PleasedPenguin
Model Tree
Qwen/Qwen3-Omni-30B-A3B-Instruct
Input Modalities
Output Modalities
Supported Functionality
python
from transformers import Qwen3OmniMoeForConditionalGeneration, AutoProcessorĀ model = Qwen3OmniMoeForConditionalGeneration.from_pretrained( "PleasedPenguin/A2R-30B-A3B", dtype="auto", device_map="auto")processor = AutoProcessor.from_pretrained("PleasedPenguin/A2R-30B-A3B")bash
vllm serve PleasedPenguin/A2R-30B-A3B --max-model-len 32768A2R reasons before it answers, and closes with a structured block:
text
<reasoning> ā¦ </reasoning><answer>{"Answer": "B"}</answer>Parse the last <answer> block. Scanning the whole string will pick up option letters quoted
inside the reasoning trace.
Multiple audio inputs are presented as a single waveform, with any reference voices appended after the main clip. The evaluation harness assembles this for you.
bibtex
@misc{lee2026hearsaidwhatunlocking, title = {HEAR Who Said What: Unlocking Speaker-Attributed Reasoning via Counterfactual Voice Grounding}, author = {Dongwook Lee and Sangkwon Park and Eunwoo Song and Che Hyun Lee and Youngho Cho and Junho Kim and June Young Yi and Heeseung Kim and Sungroh Yoon}, year = {2026}, eprint = {2608.29120}, archivePrefix = {arXiv}, primaryClass = {cs.CL}, url = {https://arxiv.org/abs/2608.29120}}Apache 2.0, inherited from Qwen/Qwen3-Omni-30B-A3B-Instruct.
@misc{lee2026hearsaidwhatunlocking, title = {HEAR Who Said What: Unlocking Speaker-Attributed Reasoning via Counterfactual Voice Grounding}, author = {Dongwook Lee and Sangkwon Park and Eunwoo Song and Che Hyun Lee and Youngho Cho and Junho Kim and June Young Yi and Heeseung Kim and Sungroh Yoon}, year = {2026}, eprint = {2608.29120}, archivePrefix = {arXiv}, primaryClass = {cs.CL}, url = {https://arxiv.org/abs/2608.29120}}