Extropic's Z1T hits GPT-2 level on sparse probabilistic hardware with up to 140x efficiency
beffjezos · x · 2026-09-05
Extropic launched Z1T, its first family of transformer-like models built for sparse probabilistic hardware, with founder Beff Jezos (Guillaume Verdon) explaining the significance.
- Claims up to 140x energy efficiency gains over GPUs and reveals a new scaling law for sparse transformers
- A year ago the first result was MNIST-level ("the 80s of machine learning"); today GPT-2-level models run on Extropic's own hardware
- Jezos argues this proves arbitrarily scalable models are possible on a new hardware substrate, and future work may push even more layers onto the hardware
- If 1000x inference efficiency gains arrive soon, he claims a multi-trillion-dollar economic impact
Related event: Extropic Unveils Z1T, Claiming Up to 140x GPU Energy Efficiency(7 posts)→
More from Infra
- Corporate token spending growth collapses from 42% to 0.2% as hyperscaler ROIC falls to 20% — SumitGup · 2026-10-21
- Qualcomm plans worldwide performance-based layoffs as it chases $5B AI data center revenue — rohanpaul_ai · 2026-10-21
- vLLM at PyTorchCon 2026: KV cache and kernel optimizations across TPU, AMD, and NVIDIA — PyTorch · 2026-10-21
- antirez to port DeepSeek Infer kernels to DGX Spark's SM121 for DwarfStar — antirez · 2026-10-21
- Ternary-Bonsai-2-9B: 2-bit ternary vision-language model trends on Hugging Face for on-device use — prism-ml · 2026-10-21
- General Compute x Positron: GPUs Handle Prefill, Memory-First Chips Tackle Decode-Bound LLM Inference — cneuralnetwork · 2026-10-21