Observed01
Reported in Arena
Community testers reported the mona-lisa-1 name and published images from their Arena sessions. This confirms the observation, not who developed the model.
Mona Lisa 1Community-observed. Not yet official.
Mona Lisa 1 is an unannounced image model name reported in Arena. Community testers suspect a connection to OpenAI, but OpenAI has not confirmed the model, its developer, or a release.
Independent tracker
Not affiliated with OpenAI or Arena. Updated as the model's status changes.
A closer look at the early results
Explore community-reported Mona Lisa 1 outputs with the original author, prompt, source, and evidence level preserved.
Geometrically consistent mirror
Lucas van de Runstraat · Arena observation
Prompt
A person stands in front of a large mirror cracked into 18 irregular pieces, with none missing. Each fragment remains in the original mirror plane but reflects a different portion of the same person and room according to its exact position. The combined fragments must reconstruct one geometrically consistent reflection rather than 18 independent portraits.
Bicycle drivetrain under load
Lucas van de Runstraat · Arena observation
Prompt
Extreme macro photograph of a bicycle drivetrain under heavy pedaling load. Show the chain engaging correctly with individual teeth on the front chainring and rear cassette. Chain links must maintain constant pitch and orientation around both sprockets, with realistic tension difference between the upper and lower chain runs. No merged links, floating rollers or impossible tooth engagement.
Seven-line metro map
Lucas van de Runstraat · Arena observation
Prompt
Design a fictional metro map containing exactly 7 lines and 42 uniquely named stations. Every line must have two endpoints. Exactly 9 stations are interchanges, including one station where four lines meet. Below the map include a complete alphabetical station index whose names correspond exactly to the stations shown on the map. Clean professional transit-design typography, no duplicate station names.
Community example, image 3
wojo · Arena observation
White's Arena example, image 1
White · Arena observation
White's Arena example, image 2
White · Arena observation
Community-sourced observation, not an official release sample.
Mona Lisa 1 is not publicly available
Get a verified status update if that changes, or create now with GPT Image 2.
What is confirmed, and what is not
Here is what has been observed, why people suspect an OpenAI connection, and what remains unconfirmed.
Speculated02
Possible OpenAI connection
Some testers describe it as a new GPT Image model based on the comparison context and perceived output similarities. That attribution remains community speculation.
Not confirmed03
No official release
OpenAI has not announced Mona Lisa 1. There is no official model page, public API model ID, release date, or confirmed relationship with GPT Image 2.
Mona Lisa 1, side by side
A closer look at the model comparisons shared by the community.
Mona Lisa 1 is not publicly available
Get a verified status update if that changes, or create now with GPT Image 2.
What stands out in the early results
The early sample set highlights four areas worth watching. These are observations to test against more evidence, not verified model capabilities.
Natural image texture
Early comparisons draw attention to ordinary skin, sensor noise, imperfect framing, and unpolished light instead of a uniformly synthetic campaign look.
Dense scene control
Dense scenes provide a useful test of signs, vehicles, storefront details, reflections, and requested text appearing in the right place.
Prompt fidelity
Constraint-heavy prompts show which requested details survive the full image and where plausible but incorrect substitutions still appear.
Visual consistency
Repeated tests can reveal whether the requested photographic language holds across outputs or shifts enough to change the creative decision.
See what image generation can do
Explore diverse scenes and creative applications generated with GPT Image 2 to experience the model's range, precision, and creative potential.
Mona Lisa 1 FAQ
Direct answers about the model name, its possible OpenAI connection, release status, sources, and current availability.
What is the Mona Lisa 1 model?
Mona Lisa 1 is an unannounced image model name reported by community testers in Arena. Its developer, final identity, capabilities, and release status have not been officially confirmed.
Is Mona Lisa 1 developed by OpenAI?
It may be, and several community posts explicitly make that attribution. However, this remains community speculation. OpenAI has not announced Mona Lisa 1 or confirmed that it operates the model.
Why does the community think it could be an OpenAI model?
The speculation comes from how some Arena tests label and compare the model, plus perceived similarities to the GPT Image line. Output similarity and community labeling are not enough to prove ownership.
Can I use Mona Lisa 1 here?
No. There is no public Mona Lisa 1 endpoint on this site. The separate Create workspace uses the publicly available GPT Image 2 service and labels that model directly.
When will Mona Lisa 1 be released?
There is no confirmed release date. The waitlist only sends source-backed status changes and does not promise access or timing.
Are the displayed images official samples?
No. Arena images on this site are community-sourced observations. Each published item must retain its author, original post, prompt when available, and evidence level.
Is this an official OpenAI or Arena site?
No. mona-lisa-1.app is an independent model watch and image-creation product. It is not affiliated with OpenAI, Arena, or another model provider.
What powers the Create workspace?
The working image generator uses GPT Image 2. Mona Lisa 1 is not offered as a model option.
Follow the evidence. Keep creating.
Get notified when the model's verified status changes, or use the publicly available GPT Image 2 model now.