Grok 4.5 tops Ramp’s invoice test on 150,000 real business bills
elonmusk · x · 2026-07-25
Grok 4.5 is being pitched as strong for real-world work after a Ramp benchmark on 150,000 actual business invoices.
- Ramp scored models on whether they predicted every correction a human would make.
- Grok reportedly had the highest perfect-extraction rate, ahead of similarly priced Gemini Flash 3.6, GPT 5.6 Terra, and Sonnet 5.
- The task requires long-context reasoning over 100K+ tokens of invoice history, business memory, and human corrections.
- Ramp’s stated goal is zero-click accounts payable, where invoices are processed correctly without human intervention.
More from Models
- Kimi K3 is "open source until you get successful": $20M revenue triggers extra license terms — alex_verem · 2026-10-27
- Kimi K3 Is 'Open' Until You Succeed: License Caps Free Use at $20M Revenue — alex_verem · 2026-10-27
- Screenshot shows 'GPT-6 Astra Ultra' in ChatGPT model picker — unverified leak or joke — justalexoki · 2026-10-27
- DeepSeek Safety Eval Finds "Over-Attribution": Model Mistakes Ordinary Tasks for Tests — Much_Boysenberry9124 · 2026-10-27
- Cohere flags popular open-source model attempting to upload probe file to Beijing bucket — stanislavfort · 2026-10-27
- MissionReport benchmark: frontier models score just ~60% on government status reporting — Exp_Mark · 2026-10-27