HN Explorer

bazlightyear

↗ HN profile [ 62.1 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. AI code looks clean. That's the Trap (thinkpol.ca)
    1 points 1
  2. Kimi K2.6 just beat Claude, GPT-5.5, and Gemini in a coding challenge (thinkpol.ca)
    380 points 57
  3. Claude vs. Gemini vs. ChatGPT vs. Grok (boreal.social)
    1 points 1
  4. Show HN: GreedyPhrase – 1.21x better compression than GPT-4o tiktoken, 6x faster (github.com)

    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…

    1 points 0