Inkling from Thinking Machines generates visuals as code, sparking neuro-symbolic debate
_Hao_Zhu · x · 2026-11-09
Blogger ZhongRuiqi shared hands-on experience with Inkling from Thinking Machines, calling visual creation with it "seriously fun" and arguing that the future of image/video generation may not be pixels but code — with neuro-symbolic AI turning out to be the answer, despite the term being previously mocked as "grumpy academia."
The quoted original post shows Inkling 2 Small can "paint" (video + audio made in code), with side-by-side comparisons against GLM 5.3 Flash, DeepSeek V4.1 Flash and Qwen 3.8 Flash on the same prompts.
More from Multimodal
- Seedance 2.5 generates 30-second single-shot factory flythrough from one prompt — LudovicCreator · 2026-11-09
- Midjourney SREF share: style code 2950214755 for eerie red-toned surrealism — tisch_eins · 2026-11-09
- AI-generated samurai remake of 90s cartoon Recess battling over giant sushi — LudovicCreator · 2026-11-09
- One Veo 3 ASMR glass-slicing video hits 66.1M views and earns $8,570 — Aiden_Tech_Ai · 2026-11-09
- Stunning vortex ring AI wallpapers with particle-level detail go viral on X — Kyrannio · 2026-11-09
- Testing AI character consistency for a virtual artist's music video — NoraVelenzaMusic · 2026-11-09