Opus 5 is said to generate much stronger visuals, with a standout wind-tunnel demo
cedric_chee · x · 2026-07-25
The post says Opus 5 can produce much stronger visual outputs, and calls out a wind-tunnel demo as especially impressive.
The attached image shows a rendered red car in a simulation scene with airflow visualization, suggesting the model or tool can generate or manipulate fairly complex visual content rather than only text. The post is brief, but the demo itself is the main information.
Related event: Anthropic's Opus 5 Reportedly Launches with Enhanced Visuals(2 posts)→
More from Multimodal
- Moonshot AI launches Kimi K3 with 2.8T parameters and a 1M-token context — dl_weekly · 2026-07-25
- Midjourney prompt shares a risograph look with grain, misregistration, and halftone dots — tisch_eins · 2026-07-25
- Midjourney style exploration leans into dark cinematic portraiture — Salmaaboukarr · 2026-07-25
- Qwen3-VL 4B in ComfyUI shows zero-token bugs when thinking is off — audax8177 · 2026-07-25
- Genfocus’s Flux1-dev refocus workflow keeps crashing on a 3090 — ReasonablePossum_ · 2026-07-25
- Snap tests in-message song generation, echoing the AI-music race — venturetwins · 2026-07-25