Elorian launches visual reasoning model Glimpse alongside DecoyCount benchmark
AndrewDai · x · 2026-10-30
Visual reasoning startup Elorian (emerged from stealth in April with a $55M seed co-led by Striker Venture Partners, Menlo Ventures, and Altimeter) launched Glimpse, pitched as a new category of visual thinking model that reasons about the physical world from images. With AI data company Mercor it also released the DecoyCount benchmark, testing whether models genuinely count objects or rely on visual shortcuts and substitution heuristics.
- CEO Andrew Dai discussed the results and positioning on Latent Space
- Fun detail: their agent accidentally published the model before the benchmark, reversing the planned order
- Core question: frontier models can see but struggle to reason past decoys; Glimpse aims to bridge that gap by thinking with images
More from Models
- ML Commons launches 'benchmarks as a service' to kill benchmaxxing — RexDouglass · 2026-10-30
- Gemini 4 Carbon takes #1 on Artificial Analysis, Anthropic S-1 in 60+ charts — altryne · 2026-10-30
- Leaker shows OpenAI shipped his ChatGPT texting feature ideas two days later — flowersslop · 2026-10-30
- First true omni model runs fully in-browser: 280 tok/s with WebGPU — victormustar · 2026-10-30
- Hidden ChatGPT checkout text reveals shared usage limits across ChatGPT and Codex starting December — btibor91 · 2026-10-30
- Grok 4.7 ranks least sycophantic model in AA-PersonalityBench, Slop score falls from 68 to 6 — XFreeze · 2026-10-30