Thinking Machines trains 594B Inkling 2 Small in 68 hours with native 4-bit
jonlachman · x · 2026-11-14
Thinking Machines details how it pretrained the 594B Inkling 2 Small in just 68 hours on GB300 NVL72 racks using NVIDIA NVFP4, with a native 4-bit forward pass from the first pretraining step through the final RL rollout — the largest known end-to-end 4-bit-trained model. The recipe yields an 8% pretraining and 20% RL wall-clock speedup over full MXFP8 while eliminating quantization calibration. NVIDIA AI publicly congratulated the team.
Related event: Thinking Machines Details End-to-End 4-Bit Training of Inkling 2 Small(10 posts)→
More from Infra
- SemiAnalysis: Google TPUs match Nvidia in nightly benchmarks; the moat is Jensen's supply chain — elonmusk · 2026-11-14
- Musk warns Nvidia's handshake supply deals may unravel without Jensen — elonmusk · 2026-11-14
- Report: Nvidia veterans may bolt post-Jensen, most already financially free — basedjensen · 2026-11-14
- Jensen Huang calls TSMC chief directly over GPU batch delays, sparking org debate — IgorCarron · 2026-11-14
- Cisco logs strongest revenue growth in 15 years, credits agentic AI demand — Justgototheeffinmoon · 2026-11-14
- NVIDIA and TSMC reportedly never signed a formal contract in nearly 30 years of partnership — IgorCarron · 2026-11-14