Per-task costs span 20x: GPT-6.2 Sol at $2.63 vs Claude Fable 5.2 at $54.25
haider1 · x · 2026-11-18
- The author observes that OpenAI keeps improving model intelligence while cutting both token consumption and cost with each release, and argues OpenAI is positioning as the cost-effective workhorse for enterprise workflows while Anthropic pushes the frontier in coding.
- The attached Artificial Analysis charts compare total cost and output tokens per task: GPT-6.2 Sol is just $2.63 (94k output tokens) and GPT-6.2 Astra $2.98, versus Claude Fable 5.2 at a whopping $54.25 (577k tokens) and Claude Opus 5.5 at $22; Gemini 4 models land between $5.83–$14.35.
- The data is broken down by reasoning effort levels (med/high/xhigh/max), showing how each vendor's cost curves scale with thinking budget.
More from Models
- Raschka: benchmarking new LLMs is like using IQ tests for hiring — rasbt · 2026-11-18
- Open-source Karpo-Decision-31B tops 300-model Decision Index with zero-token, one-forward-pass answers — rohanpaul_ai · 2026-11-18
- GPT-6.2 backlash is real user experience, not astroturf: dev says Astra has quirks but Sol is a solid daily driver — pvncher · 2026-11-18
- LLM protein-folding test on fresh PDB entry lands near-random structures — MoAlQuraishi · 2026-11-18
- Every's founder pushes back: we re-tested the model to the wire, OpenAI gave us no heads-up — kieranklaassen · 2026-11-18
- HF trending: Thinking Machines' Inkling-2-Small, MiniMax H3-Image and Qwen3.8-27B surge together — victormustar · 2026-11-18