DeepSeek V4 Test Cases Leak, Draw Criticism
teortaxesTex · x · 2026-07-18
A user shared suspected DeepSeek V4 test cases: it generated an FPV flight simulator based on SO(3) attitude control in one go, including hardware mapping, dynamics calculations, a PID controller, and rendering. However, community expert teortaxesTex criticized the V4 test cases for their mediocre visual quality, predicting that while it will be smarter and cheaper than Kimi K3, its aesthetic experience might be exceptionally poor, though K3 could potentially make up for this visual shortcoming.
More from Models
- Gary Marcus says LLM math skills are like knowing only a car’s engine size — GaryMarcus · 2026-07-22
- OpenAI’s Codex + GPT-5.6 Sol hits 99% recall in Project APE verification tests — soumitrashukla9 · 2026-07-22
- OpenAI rolls out voice in GPT-Live, but the UI obscures search and reasoning — Graham_dePenros · 2026-07-22
- Moonshot’s Kimi K3 sets a new open-weights ECI record at 156 — scaling01 · 2026-07-22
- Nanbeige4.2-3B launches as a 3B Looped Transformer model that beats larger baselines — Wooden-Deer-1276 · 2026-07-22
- A post says six companies now beat Google’s best LLM, including two open-source models — soham_btw · 2026-07-22