TSU Protocol: Open-Source RISC-V NPU for Edge AI
The Problem
AI inference needs dedicated hardware, but existing options are expensive and proprietary. NVIDIA's Grace Hopper costs $30K+. Apple's Neural Engine is locked to macOS. Qualcomm's DSP requires licensing.
TSU Protocol is building the open alternative.
Architecture
RISC-V RV64 + 16 custom Agent-extended instructions:
- MatMul & Attention — hardware ops for transformer models
- Softmax & RMSNorm — normalization in silicon
- Agent Secure Enclave — hardware-isolated agent execution
- Mesh Network — on-chip scaling
| Tier | Power | Precision | BOM | Target |
|---|---|---|---|---|
| TSU-M1 | 5W | INT8 | $150 | Edge/IoT |
| TSU-M2 | 20W | FP16/INT8 | $300 | On-device AI |
| TSU-M3 | 45W | FP16/BF16 | $550 | Enterprise edge |
Open Source. Community-Funded.
Everything is open: ISA spec, Verilog RTL, microarchitecture. No NDA. No royalties.
Current Status
Seeking $50K-$200K in community funding to cover our first MPW tape-out on 28nm/22nm. All funds DAO-governed — released transparently on milestone votes.
💰 Sponsor TSU Protocol
If you believe in open-source AI hardware, your contribution directly enables our first tape-out.
USDT (TRC-20): TU8NBT5iGyMNkLwWmWmgy7tFMbKnafLHcu
BTC: bc1ph7qnaqkx4pkg4fmucvudlu3ydzgwnfmxy7dkv3nyl48wwa03kmnsvpc2xv
Seeking **$50K-$200K in community funding to cover our first MPW tape-out on 28nm/22nm. All funds DAO-governed — released transparently on milestone votes.**
GitHub: https://github.com/JesesePU/tsu-protocol — Web: https://landing-ivory-theta.vercel.app
Anonymous. Open-source. DAO-governed. No company, no VC — just code, community, and silicon.
Top comments (0)