GPT Image 2.6 Flare Generates Nearly Twice as Fast, Scores Higher on Both Benchmarks
ArtificialAnlys · x · 2026-11-10
Artificial Analysis' independent benchmark shows GPT Image 2.6 Flare generates images almost twice as fast as GPT Image 2.5 Flare.
- Median generation time at max quality: about 30 seconds for 2.6 Flare vs about 55 seconds for 2.5 Flare.
- It also scores higher on both AA-Image-T2I v2.1 and AA-Image-Editing v2.1.
Related event: GPT Image 2.6 Doubles Generation Speed, Tops AA Benchmarks(2 posts)→
More from Models
- Dev's early take on Fable 5.2: "best model," competitors will struggle to catch up — daniel_mac8 · 2026-11-10
- A casual Monday in AI: Fable 5.2 tops AA index at 65, seven major stories in one day — Hesamation · 2026-11-10
- 6 of 7 Every testers rate Claude Fable 5.2 gold: great for long-running handoff projects — every · 2026-11-10
- System card reveals METR tested a stronger internal Claude snapshot than the released model — peterwildeford · 2026-11-10
- METR doc confirms Fable 5.2 is 'Model 3 Early', hinting Anthropic holds stronger unreleased models — peterwildeford · 2026-11-10
- Thinking Machines' Inkling 2 Small hits OpenRouter free, 594B MoE with 1M context — soumithchintala · 2026-11-10