AndreSlavescu

6 ms·

1. 1.▲
   Speech to Speech Qwen3-Omni visualization tool(qwen3-omni-mechanics-gui-website.vercel.app)

   9 points by AndreSlavescu 9mo ago|1 comments
2. 2.▲
   by AndreSlavescu 9mo ago

   Hey everyone! We at Hathora have recently released our ultra-low latency deployment of Qwen/Qwen3-Omni-30B-A3B-Instruct, one of the leading open source speech-to-speech-capable models. Platform release: https://models.hathor
3. 3.▲
   by AndreSlavescu 9mo ago

   At the moment, no unfortunately. However, to my recent knowledge of open source alternatives, the vLLM team published a separate repository for omni models now: https://github.com/vllm-project/vllm-omni I have not yet
4. 4.▲
   by AndreSlavescu 9mo ago

   From my understanding of the above problem, this would be something to do with the model weights. Have you tested this with the transformers inference baseline that is shown on huggingface? In our deployment, we do not actually tune the mod
5. 5.▲
   by AndreSlavescu 9mo ago

   Yeah, that's something we currently support. Feel free to try the platform out! No cost to you for now, you just need a valid email to sign up on the platform.
6. 6.▲
   by AndreSlavescu 9mo ago

   We actually deployed working speech to speech inference that builds on top of vLLM as the backbone. The main thing was to support the "Talker" module, which is currently not supported on the qwen3-omni branch for vLLM. Check it ou