Family dinner in China in 2005
Four family members remain around the same dinner table as the father lifts food and the mother nods

How to use
Use the exact recorded prompt. No reference image is required. Changes to the prompt or inputs create an untested variant.
Observed result
Four family members remain around the same dinner table as the father lifts food and the mother nods. Old television and 2005 calendar support the fictional period scene. Six-frame composed GIF, not historical footage.
GIF assembly
The model generated a 3-column by 2-row PNG sprite sheet. Crop equal cells in row-major order and encode six frames at 300 ms per frame with infinite looping. The resulting GIF is 512 × 512 pixels and 1.8 seconds per cycle. No interpolation was added. This is a composed animation, not a native GIF returned by the model.
Source and generation
The source provides an original prompt; this production uses the separately recorded working prompt. Source. Source-post engagement does not measure this newly generated example.
Generated with GPT Image 2.6. All reference images used here were generated for this project.
Test prints
0230-A
- Model
- GPT Image 2.6
- Recorded
- Nov 4, 2026
- Provider
- OpenAI
What we saw: Four family members remain around the same dinner table as the father lifts food and the mother nods. Old television and 2005 calendar support the fictional period scene. Six-frame composed GIF, not historical footage.
GIF
Source and licence
- Prompt
- By Prompt SeenLicence: CC BY 4.0
- Image
- Creator: Prompt SeenLicence: CC BY 4.0Generated image
Inspiration & source evidence
This prompt is an independent adaptation. The image shown here is our own recorded result.
- Original content
- X · Jason 数字生活
- Source snapshot
- Original source post; not this new generated output. Multiple cases from one post share the metrics.Likes: 1 · Reposts: 1Observed 2026-11-04
- Evidence
- Prior research: direct page and visual review. Counts have not been refreshed.The source provides an original prompt; this production uses the separately recorded working prompt.
These figures belong to the original post or video. They are dated snapshots, not live counters or ratings of our image.