Qwen4-Max launches: hands-on test shows solid spatial understanding, thrifty thinking tokens
karminski3 · x · 2026-11-16
Blogger karminski3 tested the newly released Qwen4-Max by feeding it original photos/videos and asking it to recreate scenes in three.js.
Per the comparison screenshots, the model's spatial understanding is notably improved — it accurately reproduced keycaps and control-panel button layouts, showing solid VLM capability. The three.js output isn't visually flashy, which the author attributes to limited post-training on the library, though this also keeps hallucinations restrained. Its thinking phase is also quite token-efficient, jumping to the answer quickly. A detailed test will follow.
More from Models
- Leak: Grok bots are gaining call-handoff to hand live bot calls to another number — nima_owji · 2026-11-16
- Fine-tuned 68M model beats Claude Opus 5 on bounded tasks at 12ms on CPU — sudalairajkumar · 2026-11-16
- Translated open-sources Lara Mini Base: 0.6B-active MoE beats models trained on 4-6x more tokens — NielsRogge · 2026-11-16
- Gemini 4 Hydrogen launched: 2.5x faster than 3.8 Flash, but how much is batching? — krishnan · 2026-11-16
- X User Flags Likely AI-Slop Blog That Repackages Standard Model Ladder Training as Novel 'Laddering' — code_star · 2026-11-16
- 'Alignment of the Gaps': Higher Eval Scores Don't Mean Newer Models Are Truly More Aligned — gleech · 2026-11-16