Internal eval: Grok-4.7 ranks #3 on 22 knowledge-work tasks at under $5 total cost
himanshustwts · x · 2026-09-22
The author evaluated Grok-4.7 on an internal sample of 22 knowledge-work tasks spanning engineering, bizops, finance and healthcare. It ranked #3 on the table at a total cost under $5, showing strong multi-document policy reasoning and precedence handling, performing well on large structured audits across CSV, JSON, Markdown and spreadsheets, and consistently producing the requested deliverable files. The RT quips that DeepSeek V4.1 Flash still leads on this eval, crediting Colossus 2 and Cursor data.
Related event: Grok-4.7 Ranks Third Across 22 Knowledge Tasks, Costs Under $5(2 posts)→
More from Models
- ChatGPT now links to brand homepages far more, reshaping brand-mention tracking — lilyraynyc · 2026-11-12
- YC-backed Monte launches post-training platform to build proprietary models from your agents' real work — ycombinator · 2026-11-12
- Leak Claims OpenAI Will Launch Entire GPT-6.2 Family Tomorrow to Counter Opus 5.5 and Fable 5.2 — heypearlai · 2026-11-12
- ARC-AGI-3 leaderboard: models score 90%+ but at $10K+ per task — haider1 · 2026-11-12
- LMArena Launches Agent Arena for Head-to-Head Agent Evaluation — arena · 2026-11-12
- LLMs lack hyperlocal municipal data, explains disputed Texas power outage debate — csuwildcat · 2026-11-12