Grok 4.5 looks unusually token-efficient in CursorBench 3.2
zainhas · x · 2026-07-28
The poster says Grok 4.5 looks unusually token-efficient, based on a comparison between the model’s medium and high settings.
The attached CursorBench 3.2 chart shows Grok 4.5 high and medium clustered near the top of the score range while using far fewer output tokens than several competing models, which the poster reads as a strong efficiency win.
Related event: Grok 4.5 Token-Efficiency Claims Draw Scrutiny(2 posts)→
More from Models
- Thrive Holdings launches real-work AI evals from 75+ businesses, benchmarking 13 models — samaysham · 2026-10-16
- Microsoft's MAI-Transcribe now handles 61% of all transcription on OpenRouter — mustafasuleyman · 2026-10-16
- Reflection AI Releases Open-Weight 501B MoE Beam for Agentic Coding on AWS — AWS ML Blog · 2026-10-16
- Dev RL-tunes Qwen 27B against slop with Meta's XAR method, lands top-4 on EQbench — Mysterious_Hearing14 · 2026-10-16
- HotDogBench tests 10 LLMs on 1,000 hot dog photos — all within each other's error bars — josh_wills · 2026-10-16
- TabPFN sweeps Kaggle: now in 17% of public notebooks and 3 of September's top-10 solutions — ph_singer · 2026-10-16