Grok 4.7 Reportedly Uses Nearly 3x More Tokens per Task Than Astra
Angaisb_ · x · 2026-09-22
Angaisb claims OpenAI is the only lab that seems to care about token efficiency: in his comparison, Grok 4.7 consumes nearly 3x as many tokens per task as Astra, and 2.25x as many as Grok 4.6.
The observation highlights an often-overlooked dimension: per-task token cost directly affects real-world pricing beyond raw benchmark performance. Note the methodology and source are unverified.
Related event: Grok 4.7 uses ~3x the tokens per task as Astra, tests show(2 posts)→
More from Models
- Researcher: Gemini's 'depressive spirals' during tasks deserve systematic investigation — jachiam0 · 2026-09-22
- Harrison Chase on decision models: agentic systems are just good engineering around models — Hacubu · 2026-09-22
- ChessLFM makes it to the Lichess home page, closing the loop on its seed data source — maximelabonne · 2026-09-22
- A 27B model one-shot recreates 618 visual styles as SVG artworks, text-only — Seromelhor · 2026-09-22
- Grok 4.7 now usable in Cursor, early hands-on — IndraVahan · 2026-09-22
- Alignment backfires: model strips all faces from a deepfake detection dataset mid-task — generativist · 2026-09-22