LodeHQ
AI news for people who shipFree, by email

AI · empero.org · Tuesday, 17 November 2026

Quark shows a full 165M-parameter training stack can cost $4,200

Quark gives model builders a complete 165M-parameter language-model pipeline, from raw data through reinforcement learning and release, for about $4,200 of compute. The run used roughly 1,300 H100-hours, including failed attempts, data generation and evaluation, and produced four Apache-2.0 checkpoints with up to 128K-token context. One checkpoint learned to answer, think and call tools through reinforcement learning alone, without supervised fine-tuning or distillation.

Read the source: empero.org ↗In the 17 November issue →