aivs.tv

AI Coding

Mario World 1-1 in code, Sol vs Opus — GPT-6.1 Sol vs Claude Opus 5.5

0:00 / 0:11
Video bychacha@maker_evan · originally posted on X
View original post on X

0 likes7 views

Scoreboard2

  1. GPT-6.1 SolOpenAI
    Author’s pick
    Time
    4.5 min
    Tokens
    99K
    • Top
  2. Claude Opus 5.5Anthropic
    Time
    15 min
    Tokens
    1.5M
    • Bottom

Prompt

Mario World 1-1 in code, Sol vs OpusThe post doesn’t quote the prompt wording.

Author’s verdict
The author calls GPT-6.1 Sol the most underrated model right now..

@maker_evan’s view (paraphrased)

GPT-6.1 Sol (top) and Claude Opus 5.5 (bottom) get the same Super Mario World 1-1 prompt, all in code. Sol shipped a playable level in 4.5 minutes with 99K tokens; Opus took 15 minutes and 1.5M tokens.

The post, itemised

Kind
Side-by-side
Task
Mario World 1-1 in code, Sol vs Opus
Category
Games
Mode
Not stated
Posted
Video length
12 s
Likes
0
Views
7
Bookmarks
0
Sponsored
No

Cost & time, as stated in the postAs stated: Sol 4.5 min, 99K tokens; Opus 15 min, 1.5M tokens.

Related

GPT-6.1 SolClaude Opus 5.5Claude Opus 5.5 vs GPT-6.1 Sol

Content belongs to its creators. Video by @maker_evan; aivs.tv hosts a copy for fast playback and links to the original post on X. Request removal

More with these models