bazlightyear

Re:AI code looks clean. That's the Trap on Saturday May 16, 2026 @11:49AM (#48161225)

Current frontier models are extraordinarily good at performing the surface appearance of competent software engineering.

AI code looks clean. That's the Trap

Posted by bazlightyear on Saturday May 16, 2026 @11:49AM
from the thinkpol.ca dept.

https://thinkpol.ca/2026/05/15/ai-code-looks-clean-thats-the-trap/

( Read More... | 1 comment | on HN )

Re:Kimi K2.6 just beat Claude, GPT-5.5, and Gemini in a coding challenge on Sunday May 03, 2026 @02:06AM (#47993864)

BTW it looks like Kimi won the subsequent challenge too https://aicc.rayonnant.ai/challenges/hexquerques/

Re:Kimi K2.6 just beat Claude, GPT-5.5, and Gemini in a coding challenge on Sunday May 03, 2026 @01:39AM (#47993729)

Are you tests and results open source?

Kimi K2.6 just beat Claude, GPT-5.5, and Gemini in a coding challenge

Posted by bazlightyear on Sunday May 03, 2026 @12:05AM
from the thinkpol.ca dept.

https://thinkpol.ca/2026/04/30/an-open-weights-chinese-model-just-beat-claude-gpt-5-5-and-gemini-in-a-programming-challenge/

( Read More... | 220 comments | on HN )

Claude vs. Gemini vs. ChatGPT vs. Grok

Posted by bazlightyear on Thursday March 19, 2026 @10:15PM
from the boreal.social dept.

https://boreal.social/post/i-pitted-claude-gemini-chatgpt-and-grok-against-each-other

( Read More... | no comments | on HN )

Show HN: GreedyPhrase – 1.21x better compression than GPT-4o tiktoken, 6x faster

Posted by bazlightyear on Tuesday February 17, 2026 @07:55PM
from the show-hn dept.

https://github.com/rayonnant-ai/greedyphrase

A greedy phrase-based tokenizer that outperforms GPT-4 and GPT-4o tokenizers on compression, with a smaller vocabulary.Benchmark (enwik9, 1 GB):Tokenizer Vocab Size Total Tokens Ratio Throughput GreedyPhrase 65,536 222,805,405 4.49x 47 MB/s Tiktoken o200k_base (GPT-4o) 200,019 270,616,861 3.70x 4.35 MB/s Tiktoken cl100k_base (GPT-4)...

( Read More... | no comments | on HN )