↑ ↓ Navigate ↵ Select ESC Close

GPT-5.2 is dumb (I’m tired of benchmarks)

2025-12-15 Science & Technology
70.3k
2.4k
465
Theo - t3․gg
Theo - t3․gg
539.0k subscribers

Unlock all features

FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.

Get My 10 Free Today

Related videos

This is bad...
27:17

This is bad...

Theo - t3․gg

104.3k views

I'm scared to make this video
23:37

I'm scared to make this video

Theo - t3․gg

187.3k views

I’m done.
45:51

I’m done.

Theo - t3․gg

185.8k views

I don’t really like GPT-5.5…
27:09

I don’t really like GPT-5.5…

Theo - t3․gg

132.8k views

This model is kind of a disaster.
38:00

This model is kind of a disaster.

Theo - t3․gg

166.2k views

I’m scared about the future of security
33:43

I’m scared about the future of security

Theo - t3․gg

61.4k views

I’m serious.
39:45

I’m serious.

Theo - t3․gg

95.7k views

gpt-5.4 is really, really good
40:21

gpt-5.4 is really, really good

Theo - t3․gg

153.7k views

I'm so f***ing tired of Obsidian.
18:29

I'm so f***ing tired of Obsidian.

Theo - t3․gg

113.3k views

Fumble of the decade
20:01

Fumble of the decade

Theo - t3․gg

80.5k views

Description

It feels like benchmarks are starting to matter less and less, because GPT-5.2 is not that good... Thank you Blacksmith for sponsoring! Check them out at: https://soydev.link/blacksmith Use code KIMI-PLS for 1 month of T3 Chat for just $1: https://soydev.link/chat SOURCES https://x.com/antonosika/status/1999961264983060752?s=46 https://www.zoom.com/en/blog/humanitys-last-exam-zoom-ai-breakthrough/ https://x.com/theo/status/1994922002386428143 Want to sponsor a video? Learn more here: https://soydev.link/sponsor-me Check out my Twitch, Twitter, Discord more at https://t3.gg S/O Ph4se0n3 for the awesome edit 🙏

Top Comments (10)

@SahilP2648 2025-12-15

Theo is swinging harder than the crypto market lmao

677 7 replies
@artrix909 2025-12-15

and... the story repeats itself again

484 9 replies
@krumbergify 2025-12-15

10: ECHO Best model ever! 20: ECHO No, it is actually kinda dumb 30: GOTO 10

322 10 replies
@MastermindAtWork 2025-12-15

Can wait for the next video to be "X-model" is the best model ever made.

178 2 replies
@bnkz 2025-12-15

"I was wrong about ___" counter: 10

133 1 replies
@briankgarland 2025-12-15

Yep. I ran back to Gemini after the first day.

72
@WewasAtamans 2025-12-15

When benchmarks score start becoming the goal, the benchmarks stop being a useful metric

35 1 replies
@kocokan 2025-12-15

The only reliable benchmark: How-I-Feel-About-It-Bench

35 1 replies
@jksilvester 2025-12-15

I've switched to Gemini after over 2 years of Claude and ChatGPT, and I don't think I'm going back to them anytime soon. Gemini is just better. It is nowhere close to what Theo is making it out to be.

22 2 replies
@yodaluca23 2025-12-15

The Writing reviewing is a really interesting bench

7

Unlock the Data Inside
Turn Videos into Knowledge

  • Get FREE 10/day: transcripts, summaries, chats
  • Chat with videos, export text & PDF
  • $1 free API credit for RAG, chatbots & research
Try it free

Free forever plan • All features unlocked

App screenshot