@maximelabonne: Ok it's crazy what you can do with text+vision > Adding vision can improve performance for text-based tasks > OCR is tr…

X AI KOLs Timeline Models
multimodal vision-language decision-models liquid-ai open-source efficiency

Summary

Liquid AI announced d1 with vision, making its first decision model capable of processing images, text, or both, with benchmarks showing it matches or beats frontier models like GPT-6.1 Sol and Claude Opo 5.5 on real-world applications while staying fast and cheap.

Ok it's crazy what you can do with text+vision > Adding vision can improve performance for text-based tasks > OCR is trivial, so you don't need any prompt engineering > It stays fast and cheap, even with extra vision inputs
Original Article
View Cached Full Text

Cached at: 10/05/26, 07:37 PM

Ok it’s crazy what you can do with text+vision

> Adding vision can improve performance for text-based tasks > OCR is trivial, so you don’t need any prompt engineering > It stays fast and cheap, even with extra vision inputs

Liquid AI (@liquidai): Announcing d1 with vision. 👁️👁️ Our first decision model now supports images, text or both as inputs. We tested d1 against GPT-6.1 Sol and Claude Opus 5.5 on six real applications, from filtering support tickets to inspecting circuit boards. d1 matches or beats GPT-6.1 Sol on

Similar Articles

Announcing d1 with vision (2 minute read)

TLDR AI

Liquid AI announced that its decision model d1 now supports images, text, or both as inputs, claiming to match or beat GPT-6.1 Sol on 4 of 6 real-world applications while costing 19x to 200x less and responding significantly faster. d1 returns yes/no, choice, or score probabilities in a single forward pass without generating tokens, answering in 200–300 ms.

Open d1: Edge decision models for text, vision, and audio (6 minute read)

TLDR AI

Liquid AI 发布了 d1 决策模型家族的两款开放权重模型 d1-3B 和 d1-omni-600M，支持文本、视觉和音频模态，通过单次前向传递直接输出答案，在 Decision Index 上超越同规模模型并可在 NVIDIA Jetson 边缘设备上实现实时推理。