Opus 5.5 / GPT-6 Sol / Luna
A Japanese summary of Simon's comparison grid.
470 gallery records and 164 timeline records. Categories may overlap and include related beagle, parrot and other experiments.
634 records · Page 2 / 27
A Japanese summary of Simon's comparison grid.
A Chinese timeline overview: unrecognizable in 2024, recognizable but incorrectly posed in 2025, static images reaching a ceiling in September 2026, then playable webpages.
The author alleged that the model traced an image into 1.3 million characters of paths rather than reasoning spatially.
An example of a two-stage Chinese prompt. The connection dropped once; the author judged its aesthetics roughly on par with K3. Video length: about 11 seconds.
Four effort levels side by side, with local file reading prohibited.
A different species. Reported cost: about CNY 8 and 4 million tokens. Published at case.ccvibe.top, with added weather, night scenes and Mario-style gameplay.
CodexRadar's “Pelican Cup” collected nearly 100 submissions to monitor possible Astra regressions.
Simon tried another prompt in one shot, without editing the code.
The source's ranking: Astra 94, GPT-5.6 Sol 88, GLM-5.3 86, Kimi K3 83, GLM-5.3-Flash 81, Sonnet 5 78, DS-V4.1-Flash 76, Qwen3.8-Max 74.
A scoring rubric assigned 25 points to pelican features, 25 to bicycle structure and 30 to animation kinematics. Astra used two-link IK.
3D workA short-prompt regression test covering instruction following, subject recognition, structure, coordinated animation, visual organization and context retention.
The classic prompt moved from SVG to a local 3D content-creation tool. Background and visual flourishes were added later.
A grid comparing reasoning effort levels. Community comments said Astra finally looked more like a bird actually riding a bicycle.