TGViewer
HN Digest | Hacker Newsʼning eng yaxshilari HN Digest | Hacker Newsʼning eng yaxshilari @hndigestuz · 75 subscribers
Post #535 28
LLM modellarining yashirin degradatsiyasini tekshiruvchi `livenerf` benchmarki
↘︎ Livenerf: Has Opus 5.5 been nerfed yet?

livenerf benchmarki Claude Opus 5.5 modeli imkoniyatlari pasayishini aniqlash uchun GPQA, MMLU-Pro va MATH testlaridan saralangan 78 ta savol yordamida 30 kunlik avtomatik tekshiruv oʻtkazadi.

💬 Foydalanuvchilar modellar sifatining pasayishini serverlar yuki ortishi yoki resurslar yetishmasligi bilan bogʻlashmoqda.
→ muhokama


857 ball · 366 izoh · #daily · @HNDigestUz
GitHub GitHub - ninjahawk/livenerf: Benchmark for tracking model capability after release. Benchmark for tracking model capability after release. - ninjahawk/livenerf
Open in Telegram All posts from @hndigestuz
More from @hndigestuz
  1. Oct 1, 2026Taqachilikdan mexaniklikka oʻtgan oilaning texnologik moslashish tarixi ↘︎ The last time m…
  2. Sep 30, 2026🔥 Gemini 4 Argon modeli 1 million chiquvchi tokenni qoʻllab-quvvatlaydi ↘︎ Gemini 4 Argon…
  3. Sep 30, 2026Doimiy ishlovchi Dots sunʼiy intellekt agentlari taqdim etildi ↘︎ Dots: Always-on agents D…
  4. Sep 29, 2026🔥 GPT 6.1 Sol modeli besh barobar arzon narxda taqdim etildi ↘︎ GPT 6.1 Sol: Near-Astra i…
  5. Sep 29, 2026AQShda uyda mehmon kutish madaniyati inqirozga uchradi ↘︎ You are no longer invited to din…
  6. Sep 29, 2026Sonnet 5.5 dasturlash testlarida Opus 5.5 darajasiga yetdi ↘︎ Sonnet 5.5 Yangi model avval…
Threads Profile ViewerView any public Threads profile without an account.Open ThreadLook →Writing with AI? Make it sound human.Metric37 rewrites AI drafts so they read naturally. Free AI detector, 1,500 words free.Try Metric37 →