The Efficiency Landscape: GPT-6.1 Sol tops value among top 30 AI models, Opus 5.5 priciest at $10.04
daniel_mac8 · x · 2026-11-05
AI blogger Daniel Mac published an "Efficiency Landscape" scatter chart comparing the top 30 AI models by Artificial Analysis Intelligence Index vs cost per task (data as of Nov 5, 2026, each model at its highest-scoring setting).
- Best value: GPT-6.1 Sol (#4 intelligence, #11 cheapest) and GPT-6.1 Terra (#2 intelligence, #10 cheapest); GPT-6 Luna sits in the cheapest tier ($0.15/task)
- Cheapest: MiMo-V2.6-Flash at $0.13; Priciest: Opus 5.5 ($10.04), Sonnet 5.5 ($10.02), Fable 5.1
- Worst value: the Qwen3.8 27B / Nemotron 3.5 Super VL region; GPT-6 Astra is #1 in intelligence (57) but costs $6.24
- Axes cross at the 30-model median (31.5 pts, $2.20), dividing models into "Quadrant of Wow" (smart and cheap) and "Quadrant of Woe" (weak and expensive)
More from Models
- Strata 0.2.11 adds Unsloth Dynamic v3.1 quants with seven 1-4bit GGUF tiers — danielhanchen · 2026-11-05
- Strata 0.2.11 ships Unsloth Dynamic v3.1 quants, 1-bit runs at 68GB — danielhanchen · 2026-11-05
- Third-Party Guide Rounds Up Everything That Changed in Claude's Major Update — blaizedsouza · 2026-11-05
- Claude falsely flagged twice: ImageNet sheep chat and securing mom's router — ducha_aiki · 2026-11-05
- Users complain Claude wastes tokens on endless apologies instead of just fixing errors — RachelVT42 · 2026-11-05
- Hy-4 vs DeepSeek Flash benchmark: quarter the tokens, a third of requests — ShunyuYao12 · 2026-11-05