Extropic unveils Z1T sparse models claiming up to 140x energy efficiency gains over GPUs
beffjezos · x · 2026-09-05
AI hardware startup Extropic introduced Z1T, its first family of transformer-like models designed for sparse probabilistic hardware like its Z1 chip. According to the company's blog, Z1T achieves up to 140x energy efficiency gains over GPUs and reveals a new scaling law for sparse transformers. This marks the company's first public release of models and benchmarks supporting its thesis that probabilistic hardware can replace conventional GPUs for generative AI.
More from Infra
- Aer Lingus says 50% of long-haul fleet now has Starlink, full coverage by year-end — elonmusk · 2026-09-05
- Running Qwen3.8-27b-mlx Locally on an M5 MacBook to Update a Website and Generate Assets — walkingriver · 2026-09-05
- Agent Substrate brings instant suspend/resume and 10x density to K8s AI agents — davemccollough · 2026-09-05
- Horizontal AI apps plus vertical hardware integration may breed dominant vendors — matt_slotnick · 2026-09-05
- Local Qwen 3.8 Flash Next benchmark jumps from 91 to 98/100 with AWQ quants and patched vLLM — WonderRico · 2026-09-05
- Coatue reportedly building multibillion-dollar JV with chip startup MatX to lock in HBM and wafer capacity — zephyr_z9 · 2026-09-05