AMD Acquires Taalas to Hard-Code AI Models Directly Into Silicon
The Decoder · rss · 2026-08-08
AMD has acquired Canadian startup Taalas, which specializes in hard-coding AI model weights directly into inference chips.
While this approach locks each chip to a single model, it delivers extreme performance. A demo chip achieved over 16,000 tokens per second per user running Llama 3.1-8B. Google is also reportedly developing a similar hardware approach for Gemini.
More from Venture
- Indie hacker tip: submitting your sitemap to GSC got 6,021 pages indexed in 2 days — alexmacgregor__ · 2026-10-16
- Grok-built linear regression bot claims $508K PnL in 125 days on Polymarket — Roger_M_Taylor · 2026-10-16
- Pay-to-rank directory outbid.lol charges $20 a day for the #1 spot, top listings paid $14k-17k — jonathan_wilke · 2026-10-16
- SignWith, a pay-per-doc e-signature tool, hits 1,000+ non-tech businesses — tibo_maker · 2026-10-16
- After 6,000 tweets and a $10M exit, tibo shares data on what works on X — tibo_maker · 2026-10-16
- Decision Models spins out its sandboxing system as a cheap standalone product — airesearch12 · 2026-10-16