LLMS

Qwen 3 release questioned technical report limitations

By

–

12 October 2025 16h01

I think the only one I might question the significance of in your list is Qwen 3. They didn't even release a base model. The technical report was quite disappointing as well.

vLLM Limitations in Broader AI Landscape

By

–

12 October 2025 10h33

vLLM is a small part of the whole picture. They're massively lacking.

→ View original post on X — @bcherny

Comparing Sonnet and Opus AI Models: Performance Differences

By

@bcherny

–

12 October 2025 9h14

Are you using Sonnet or Opus? And when you say dumber what specifically?

→ View original post on X — @bcherny

Rate Limits and Token Compaction Bugs in API Systems

By

@bcherny

–

12 October 2025 2h06

If you're seeing lower rate limits than what we publish, or if you're seeing auto-compact earlier than ~155k tokens, that's a bug. /bug to report.

→ View original post on X — @petergostev

Are We Near AGI Flight? Current LLM Capabilities Assessed

By

@petergostev

–

12 October 2025 0h15

My short essay on LLMs & AGI: Have we reached the 'AGI of flight'? Not even close. Our planes can carry hundreds of passengers across the globe in just a few hours, yet nothing we build matches a godwit flying non-stop for eleven days without rest or the agility of a bat

→ View original post on X — @skirano

Claude and GPT-5 for coding: shifting away from Cursor agents

By

@skirano

–

11 October 2025 21h41

I primarily use Claude with Claude code and GPT-5 in Codex. I rarely use cursor agents anymore.

→ View original post on X — @therundownai

Samsung’s Tiny AI Model Outperforms DeepSeek and Gemini

By

@therundownai

–

11 October 2025 21h31

5. Samsung’s Tiny AI beats the giants Samsung unveiled the Tiny Recursion Model (TRM), a 7M-parameter AI that outperformed DeepSeek R1 and Gemini 2.5 Pro on reasoning tasks using a self-improvement loop.

LLaMA 2 Sparked Open Source LLM Interest

By

–

11 October 2025 20h59

It definitely kickstarted things for LLMs but it was LLaMA 2 that showed there was interest in opensource models.

Qwen 2.5 influence research adoption compared to other models

By

–

11 October 2025 20h51

not that it's a bad model, but in terms of influence that is yet to be determined Qwen 2.5 is still used to this day in research and experiments, more than any other model – that's a different level of influence