0:00 / 0:19
Video by Furkan (@Furkan_popr) · originally posted on XView original post on X

Google Antigravity (Gemini 3.1 Pro) vs DeepSeek Harness (DeepSeek V4 Pro 0813) vs OpenAI Codex (GPT-6 Sol)

Escape-the-cops game, 3 harnesses

Furkan@Furkan_popr

Harness comparison2 likes48 views

A run-from-the-cops game from one-shot prompts in three harness and model pairs: Google Antigravity with Gemini 3.1 Pro, DeepSeek Harness with DeepSeek V4 Pro, and Codex with GPT-6 Sol.

Each side pairs a coding harness with its own model, so harness and model differences are mixed. All harness comparisons.

Harnesses and models

HarnessModelEffortNotes
Google AntigravityGemini 3.1 ProGoogleHighFree, per the author.
DeepSeek HarnessDeepSeek V4 Pro 0813medium confidenceDeepSeekHighAbout $2, per the author.Written as “V4 Pro (DeepSeek, snapshot unstated)”Written “V4 Pro”; listed as DeepSeek V4 Pro 0813.
OpenAI CodexGPT-6 SolOpenAIMediumCodex on a $20 subscription.Written as “GPT-6 Sol”

Details

KindSide-by-side (same post, same prompt)
TaskEscape-the-cops game, 3 harnesses
Author’s verdictOnly one of the three is actually playable. — @Furkan_popr’s view (paraphrased)
CategoryGames
ModeOne shot
Posted
Video length19 s
Likes / views2 / 48 · as of Oct 18, 2026
Bookmarks0
NotesThe author marks the Antigravity and DeepSeek Harness runs with a clown emoji, implying the Codex run is the playable one.
SponsoredNo
Original postx.com/Furkan_popr/status/2103138110037020798

Related

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

More with these models