Opus 5.5 vs GPT-6.1 Sol: same Blender prompt, two animations

Coding benchmarks tell you how a model does on a test. A 3D animation tells you something different: whether a model can turn a written brief into a long script that builds a scene, lights it, moves a camera, and survives a render. So we ran a simple experiment and are leaving the verdict to you.

We gave Claude Opus 5.5 (Max effort) and GPT-6.1 Sol (Ultra code) the same prompt: create a 3D animation in Blender, 900 frames, 30 seconds at 30 frames per second. The subject that came back from both was a clockwork mechanism filmed in macro. We then rendered both on the same machine, an RTX 5060, at the same time.

Watch them side by side

Opus 5.5 Max is on the left. GPT-6.1 Sol Ultra code is on the right. Each half carries its own label.

The test setup

Both renders ran at the same time on one GPU. That makes the render times a rough guide, not a controlled benchmark, and they say more about how heavy each scene was than about how fast either model works. Opus 5.5 is Anthropic's September 22 release, and GPT-6.1 Sol is OpenAI's September 29 release; each company's own launch page lists the effort levels and prices.

What we noticed

These are observations from watching the video and studying still frames sampled every two seconds. They describe what is on screen, not which result is better.

Framing. The left side stays in extreme macro. At any moment you see one or two parts, such as a balance wheel, a gear cluster, or a section of a large gear, and the subject changes noticeably across the 30 seconds. The right side shows a wider, elevated view of the whole movement on a round plate and holds that view while the camera drifts slowly closer and lower.

Depth of field. The left side uses a very shallow focus. A narrow band is sharp and much of the frame turns into soft orange and brown blur. On the right, most of the mechanism stays in focus, and the blur appears mainly toward the edges and background.

Light and color. The left has a hazy, warm, low-contrast look in gold and amber, with small blue accents from screws and jewels. The right is higher in contrast: copper against a dark grey plate, with directional light and cast shadows from the gears and pillars.

Materials. The left reads as smooth, polished brass, though fine detail is only visible in the small in-focus areas. The right shows more surface variety, with weathered copper, steel arms, a glossy silver part, and a textured plate. Gear teeth, screws, and the coiled hairspring are readable across much of the frame.

What they share. Both are a watch-style movement with a balance wheel and hairspring, both sit on a warm orange-to-terracotta background, both use slow camera movement with no cuts, and both include floating dust.

Possible weak spots. Viewers who prefer a clear view of the whole mechanism may find the left side leaves little of it visible. Viewers who like a cinematic macro feel may find the right side more of a set on a table. A few spots to check as you watch: soft areas near the upper right of the right side, parts cropped by the frame edge on both sides, and how much of each balance wheel is in focus when it is the main subject.

How to judge it yourself

Different jobs reward different things, so it helps to decide what you are choosing for before you watch again:

• Brief fit: Which animation looks more like the macro clockwork shot you would have asked for?
• Readability: Can you follow how the mechanism works, or do you mostly feel it?
• Camera and pacing: Does the movement hold your attention for 30 seconds?
• Finish: Which scene would need less cleanup before you could use it?
• Cost: Is the extra render time, and the cost of the model run behind it, worth what you see?

If you try the same experiment, keep the settings as matched as you can, render on separate runs if you want clean timings, and save the prompt so others can repeat it. We would like to hear which side you picked.