Hacker News: CapyToolkit

New comment by CapyToolkit in "The AI Aesthetic"

CapyToolkit — Fri, 31 Jul 2026 06:28:54 +0000

To me, neither of them feels like AI generated. The tarot example feels like its inspiration came from collectible card games, while the ramen example reminds me of Cyberpunk 2077.

New comment by CapyToolkit in "Gemini Robotics 2 brings whole body intelligence to robots"

CapyToolkit — Thu, 30 Jul 2026 16:13:29 +0000

I use Sonnet and Opus everyday. Believe me, it's very common they tell me that something is bad idea, write a rationale and suggest better solutions.

Gemini doesn't remember almost anything after 2-3 follow ups. I have to paste the same "system prompt" at the top of each message and it still doesn't understand it.

New comment by CapyToolkit in "Kimi-code not performing well"

CapyToolkit — Thu, 30 Jul 2026 16:06:54 +0000

Honestly, I have a similar impressions after testing every model except Sonnet, Opus or Claude. I always test them inside Claude vscode extension in the terminal, using the same skills or workflows from multiple skills.

I tested Mimo V2.5 Pro, Qwen 3.8 Max Preview, GLM 5.2, Macaron V1 Venti and many other, smaller model whenever hype on X appears. All behave dumb, don't follow instructions properly even when they write skills for themselves. They are literally on the level of Sonnet 3.5 which was released in 2024 so I use them when available on free endpoints for simple tasks like writing or subtle UI improvements...

New comment by CapyToolkit in "Gemini Robotics 2 brings whole body intelligence to robots"

CapyToolkit — Thu, 30 Jul 2026 15:43:06 +0000

If it has intelligence of Gemini... well, good luck...

New comment by CapyToolkit in "LLM Honeypot"

CapyToolkit — Thu, 30 Jul 2026 15:11:51 +0000

This is what happens when you prompt Lovable to build the website that remembers previous century standards.

New comment by CapyToolkit in "Ask HN: How are you handling production risks from AI generated code?"

CapyToolkit — Thu, 30 Jul 2026 09:30:58 +0000

If you are unsure about the AI generated code, it's better to not ship a risky product. You can also use 3rd party services for the most vulnerable areas of your product.

When Your Video Call Freezes, the Problem Is Not What You Think

CapyToolkit — Thu, 30 Jul 2026 07:07:45 +0000

Article URL: https://capytoolkit.com/blog/guides/video-call-freezes-p2p-connectivity-diagnostic/

Comments URL: https://news.ycombinator.com/item?id=49106892

Points: 2

# Comments: 0

What NPU Tops Doesn't Tell You: How Memory Bounds Local LLM Performance

CapyToolkit — Wed, 29 Jul 2026 15:34:52 +0000

Article URL: https://capytoolkit.com/blog/buying-guides/npu-tops-vs-memory-bandwidth-local-llm-ai-pc/

Comments URL: https://news.ycombinator.com/item?id=49098870

Points: 2

# Comments: 0

Wi-Fi 7 Router Buying Guide: Specs That Mean for Gaming and Streaming

CapyToolkit — Mon, 27 Jul 2026 15:57:27 +0000

Article URL: https://capytoolkit.com/blog/buying-guides/wifi-7-router-buying-guide-specs-gaming-streaming/

Comments URL: https://news.ycombinator.com/item?id=49071444

Points: 2

# Comments: 0

New comment by CapyToolkit in "Misleading AI-generated doctors pose 'huge danger to public safety'"

CapyToolkit — Mon, 27 Jul 2026 15:48:03 +0000

100% Agree. It's not only about AI-generated but also real people, often celebrities, without proper qualifications who drop generic "health tips" videos without knowing who will be recipient of their tips.

New comment by CapyToolkit in "Opus 4.8 feels worse then sonnet"

CapyToolkit — Thu, 25 Jun 2026 11:19:15 +0000

It's hard to disagree with you. If it's not enough, usage limits feel 3x less than a week ago.