Where is the LLM?

2025-11-27 07:53:01 Ask

Hi! Is the prompt sent over to Deepseek cloud model, and the assistant is only managing voice to text and text to speech, or is the AI inference happening on the device itself? Thanks!

LLM inference

APost Reply

Like0| report

2answers

SpotPearGueste93f8

Answer time:2025-11-27 09:16:29

The device mainly handles audio encoding and decoding locally and sends and receives data with the server. Model inference is not performed locally.

Like0

report

SpotPearGuest58655

Answer time:2025-12-29 12:51:20

Got it! Is there an ability to set up a different endpoint, like my local LLM server? Thanks!

Like0

report

APost Reply

DeepSeek AI Voice Robot Ball ESP32-S3 1.28 inch LCD N16R8 Development Board espHome Assistant Wifi Weather Station TimeClock Electronic EYE

Price: $16.9-20.9

Part Number： ESP32-S3-1.28inch-AI

Brand: SpotPear

Address：Rm401, building 4, Enlai industrial city, Nankang Road, Danzhutou community, Nanwan street, Longgang District, Shenzhen

Email：tech-support@spotpear.com (Only tech support .Order see 'CONTACT US' at our web )