Liquid AI's d1.2-ultra beats Google and Microsoft decision models at half the price
JosephJacks_ · x · 2026-11-19
Within two weeks, Google DeepMind and Microsoft both released decision models — yet smaller Liquid AI beat them both on the Artificial Analysis Decision Index: d1.2-ultra scored 75 vs Microsoft-Decision-1.1 at 68 and Gemini Decision 1 at 66. The DecideBench v2 chart (450 decisions, run Nov 18) shows d1.2-ultra at 418/450 correct costing $57 per million decisions, vs Gemini Decision 1 at 411/450 for $157 and Microsoft-Decision-1.1 at 410/450 for $63. Open-weights yev2-35b-a3b scored 396/450 at $95 self-hosted. Frontier reference: Claude Opus 4.8 hit 447/450 at a staggering $26,543 per million decisions.
More from Models
- Surge AI benchmarks Gemini 4 Neon: tops Tuesday@Work at 71.6, trails on SWE — echen · 2026-11-19
- Google's rumored Neon and Carbon models reportedly match or beat Opus in early tests — almmaasoglu · 2026-11-19
- Musk says Grok 4.8 is in RL phase, built for 'usefulness max' over benchmarks — XFreeze · 2026-11-19
- Prediction: Anthropic to drop Opus 5.6 tomorrow — the mysterious 'Model 4' from METR's RSI report — daniel_mac8 · 2026-11-19
- OpenAI's 96 ECI benchmarks skew to math and AI research; hard multimodal and cybersecurity tests are rare — scaling01 · 2026-11-19
- Neon author: Opus 5.5 bugs stem from over-specializing on specific test failures — neurocy · 2026-11-19