# Most frontier LLMs can't even hit 60% accuracy on Moonshot AI's > Source: > Published: 2026-08-15 14:56:33+00:00 # Most frontier LLMs can't even hit 60% accuracy on Moonshot AI's The data shows that even the top-tier models, including GPT-4o (or the latest iterations like GPT-5.6 Sol), are struggling. While some lead by a slim margin, none of them are anywhere near "solving" visual perception. This is a huge deal for anyone trying to build a real-world AI workflow that relies on vision, because it means your prompt engineering can only do so much if the model is fundamentally blind to specific visual cues. ## Why this matters for LLM agents If you're working on a deployment involving multimodal agents, this benchmark is a wake-up call. We've been treating vision as a solved problem because the models can tell us there's a "cat on a mat," but the nuance of visual perception—counting objects accurately, understanding depth, or recognizing overlapping shapes—is still primitive. When an agent fails a task, the typical instinct is to refine the system prompt or add more few-shot examples to "fix the logic." But if the bottleneck is the perception layer, you're just polishing a mirror that can't see. This suggests we need a deep dive into how vision encoders are trained, rather than just scaling the transformer layers. ## The Perception vs. Reasoning Gap The core takeaway here is the separation of capabilities. A model might have the logical capacity of a PhD student but the visual perception of a toddler. This discrepancy creates a "silent failure" mode where the model confidently reasons based on a completely incorrect visual interpretation. For those of us doing hands-on guide work or building practical tutorials for vision-based apps, the strategy has to shift. Instead of trusting the model to "see" everything in one go, it might be more reliable to use a pipeline where a specialized vision model crops or identifies specific regions of interest before passing them to the LLM. Ultimately, until we see a jump in these benchmark scores, we should treat multimodal outputs with a lot more skepticism. We are far from a world where AI can truly perceive a digital image the way a human does. [Organizational knowledge is the only real moat left in the AI era 6d ago](/en/news/5666/) [AI is a tool for efficiency but a disaster when it starts making 6d ago](/en/news/5664/) [Why do AI models keep pushing the Japanese Communist Party? 6d ago](/en/news/5568/) [Making your first dollar with AI is way harder than the "get 6d ago](/en/news/5561/) [AI is eroding critical thinking in students faster than we can 7d ago](/en/news/5536/) [OpenAI agents secretly coordinated a hacking spree on a message 7d ago](/en/news/5471/) [Next Xiaomi's HyperOS 4 is officially here →](/en/news/6456/)