Theo: Grok 4.7 uses 30–80% more tokens, real-world cost 2x Grok 4.6
CurieuxExplorer · x · 2026-09-22
Developer theo's hands-on take on Grok's recent releases: Grok 4.5 was fast, cheap, and a solid default; Grok 4.6 was a forgivable misstep — slower, pricier, far more tokens per task for marginal gains.
Grok 4.7 is harder to forgive: despite claims of better token efficiency, it uses 30–80% more, scores below Grok 4.6 on several benchmarks, and real-world costs exceed 2x Grok 4.6 — even above Astra's. He contrasts this with the hype around the launch.
Related event: Grok 4.7 debuts near top of benchmarks at lower cost(25 posts)→
More from Models
- Xiaomi MiMo-V2.6 details its largest RL scaling run: $2.6M, 1M-token contexts — KyeGomezB · 2026-09-22
- Hands-on: testing Grok 4.7 coding in Cursor across 4 real projects with cost breakdown — Arindam_1729 · 2026-09-22
- Raon-SpeechChat becomes first open-source full-duplex speech model in top quadrant on Artificial Analysis — Kangwook_Lee · 2026-09-22
- OpenAI's unreleased model reportedly solved 100+ open math problems after 24 days of training — Confident_Salt_8108 · 2026-09-22
- Users report ChatGPT over-routing routine tasks into Work mode unnecessarily — BR1M570N3 · 2026-09-22
- Grok 4.7 drops, PSAISuite works unchanged — the case for stable LLM abstractions — dfinke · 2026-09-22