In 30 seconds
Matthew Berman says he tasked GPT-6.1 Sol with recreating Super Mario World for a ModRetro handheld. After more than four days, he describes an iterative process of writing code and comparing it with the game; the project is still unfinished in the clip.
- 1The handheldBerman says he received a ModRetro device at OpenAI DevDay and wanted to play a favorite game on it.
- 2A recreation taskHe tasked GPT-6.1 Sol with recreating Super Mario World for the device and says it has worked for more than four days.
- 3An iterative processHe describes consulting online documentation, writing code and checking the result against the game. The project is not yet finished.
Takeaways
- Berman wanted to play Super Mario World on a ModRetro handheld he received at OpenAI DevDay.
- He says GPT-6.1 Sol has worked on a recreation for more than four days.
- He describes the process as writing code, comparing its appearance with the game and iterating.
- His expectation that the work might finish in a few more days is an estimate, not a demonstrated result.
Read the edited text by chapter
An unfinished game recreation
Berman describes why he started a handheld game recreation and how the AI coding process has unfolded. He estimates it may take a few more days.
At OpenAI DevDay I received a ModRetro handheld, a recreated Game Boy-style device. Super Mario World is one of my favorite games, and I wanted to play it on the handheld, although it was not made for that device.
I tasked GPT-6.1 Sol with recreating the game from scratch for ModRetro. It has been working for more than four days. I describe it as looking at how the game should appear, using documentation it finds online, writing code and comparing the result before iterating; it is not necessarily decompiling, but feels close to me.
I think it might finish in another few days. This is a progress update, not a demonstration that the recreation is complete.
More from this channel
34:21EN
Matthew Berman· Oct 27, 2026Matthew Berman Examines Anthropic’s IPO Filing, AI Growth and GovernanceMatthew Berman, Sam Altman · Forward FutureThe presenter reviews the S-1’s financial metrics, compute strategy and voting structure while questioning the tension between rapid AI development and safety.
0:36ENMatthew Berman· Oct 25, 2026Kimi K3.1 and a Rubik’s Cube Solver DemonstrationMatthew Berman · Forward FutureMatthew Berman shows a scrambled virtual cube being solved and gives the Kimi K3.1 test a favorable assessment.
25:50ENMatthew Berman· Oct 24, 2026Kimi K3.1 Pro Tested: Strong 3D Demos, Uneven Results and Competitive PricingMatthew Berman · Forward FutureMatthew Berman tests the open-weight model on coding, 3D rendering and visual explanations, then compares benchmark results and prices.
32:20ENMatthew Berman· Oct 23, 2026Who Is Local AI For? The Costs, Trade-offs and a Hybrid FutureMatthew Berman · Forward FutureMatthew Berman weighs data control and independence against hardware cost, setup complexity and inference speed.
1:03:37ENMatthew Berman· Oct 23, 2026Matthew Berman’s Team Tests Kimi K3.1 on Animation, Games, and ChartsMatthew Berman, Alex, Brian · Forward FutureA live team AMA turns into an informal test of Kimi K3.1, from personal-agent workflows to motion graphics, a browser game, and a benchmark chart.
0:53ENMatthew Berman· Oct 22, 2026Sam Altman on AI and the Discovery of New KnowledgeSam Altman · OpenAIAltman expects progress in mathematical discovery to extend to biology, materials and energy.