Decision model leaderboard called out for omitting GPT-6 Sol and Luna Decisions
gleech · x · 2026-10-12
gleech flags "chart crimes" in a decision-model comparison chart that omits GPT-6 Sol's accuracy and GPT-6 Luna (especially Luna Decisions). The chart shows Microsoft-Decision-1 leading with 83.5% average accuracy at 85ms p50 latency—2.5x faster than runner-up H2O-Lighting-4B and 35x faster than GPT-6 Sol—while Strands-Decider 2B scores just 54.8%.
Related event: Model Evaluation Report Criticized for "Chart Crimes"(2 posts)→
More from Models
- Claude Opus 5.5 Max scores 167.3 on Epoch ECI, tracking or above AI Futures trend line — SamuelAlbanie · 2026-10-12
- Users are pasting bank statements and medical reports into cloud AI — the case for local open models — alex_verem · 2026-10-12
- Dev observes models barely improved at one-shotting in 6 months — tool use and thinking are the ladder — rickasaurus · 2026-10-12
- Leaked Claude Test: One Prompt, Two Hours, a Full 3D Car Model — iamfakhrealam · 2026-10-12
- Redditor feeds 15 years of mental health crisis logs to Gemini, ChatGPT and Claude, results surprise the group — KrazyKreutz · 2026-10-12
- Prepping for Claude Pro: home broadband, $10 VPS, used router — but virtual cards fail — lxfater · 2026-10-12