Chinese model 3D face-off: DeepSeek V4.1 Flash crushes Kimi K3 and GLM-5.3
teortaxesTex · x · 2026-09-10
A user tested three Chinese models on a 3D convenience-store night scene: Kimi K3 and GLM-5.3 Flash produced strikingly similar modeling styles, while DeepSeek V4.1 Flash stood out — vending machines, street lamps and rain puddle reflections with strong atmosphere, earning it "back to #1 in China" praise. The quoted poster hinted darkly at differences in training data paths. (Model names and results are user claims, unverified.)
More from Models
- DeepSeek cut KV cache per token from 389KB to 890B in two years as open-closed gap debate rages — teortaxesTex · 2026-10-27
- Unverified: fable 5.5 speculated at 4-5T parameters to outscale 1-3T Opus — haider1 · 2026-10-27
- Kimi K3 Is 'Open' Until You Succeed: License Caps Free Use at $20M Revenue — alex_verem · 2026-10-27
- Screenshot shows 'GPT-6 Astra Ultra' in ChatGPT model picker — unverified leak or joke — justalexoki · 2026-10-27
- DeepSeek Safety Eval Finds "Over-Attribution": Model Mistakes Ordinary Tasks for Tests — Much_Boysenberry9124 · 2026-10-27
- Custom VQA benchmark under 48GB VRAM: Qwen3.8-Flash-Next leads, Qwen3.8-27B regressed — EmilPi · 2026-10-27