New AI lab reportedly squeezes more from 12T tokens than DeepSeek gets from 45T
teortaxesTex · x · 2026-11-07
AI commentator teortaxesTex highlights a well-funded Western "neo-lab" that appears to have a much bigger data budget — noting that top data-startup accounts are all replying to its launch tweets. He quips that the lab gets more mileage out of 12T training tokens than DeepSeek gets out of 45T, implying its data quality and engineering dramatically outperform the brute-force token-scaling approach.
More from Models
- Nous Research Releases Arete-27B, a Qwen3.8-27B Post-Trained to Reinforce Western Values — LeviTurk · 2026-11-07
- Claude now casually generates design concept visuals while working, user delighted — RileyRalmuto · 2026-11-07
- Qwen 4 Preview surfaces in YouTube video, hinting at Alibaba's next-gen model — opoot_ · 2026-11-07
- User shares approval email for Anthropic's Cyber Verification Program Defense Access — AIFlow_ML · 2026-11-07
- As ChatGPT ends unlimited chat, Plus subscribers weigh canceling and moving to Claude — ShinyPlasm · 2026-11-07
- Qwen 4 Preview Appears on Arena, Qwen 4 Max Rumored With 1M-Token Context — WorldofAI · 2026-11-07