Samuel Zeng
@SamuelZengML
Founder, | On-device foundation models. High intelligence, low memory, low power | MIT TR35
235 Following 3.7K Followers
Something big is coming on September 10. A major open-source release from us.
35B parameters. One iPhone. No cloud.
We trained Edge8-35B, an ultra-sparse MoE with a jointly trained dynamic expert planner, and built an SSD-streaming inference engine around it.
In this demo: 44 tok/s, ~1.06 GB peak memory.
A truly usable large-model stack for on-device AI.
Model, runtime, and paper: open source soon.
Show more
Audio8 TTS Preview 0.1B — now on Hugging Face as ONNX INT8
~100M params · 11 languages · zero-shot voice cloning
Runs on CPU only (~0.4 GB RAM) — no PyTorch, no CUDA
44.1 kHz · OpenAI-compatible API · Apache 2.0
Try it 👇
#TTS# #VoiceCloning# #ONNX# #OpenSource#
Show more
35B parameters. One iPhone. No cloud.
We trained Edge8-35B, an ultra-sparse MoE with a jointly trained dynamic expert planner, and built an SSD-streaming inference engine around it.
In this demo: 44 tok/s, ~1.06 GB peak memory.
A truly usable large-model stack for on-device AI.
Model, runtime, and paper: open source soon.
Show more
Audio8-ASR-0.1B delivers real-time Speech-to-Text in 7 languages:
English, Chinese, Cantonese, French, German, Japanese, and Korean.
Low-latency on-device inference with ONNX Runtime + Apple Neural Engine.
Model:
Show more
