Hackernews posts about Benchmarks

1. GLM 5.2 beats Claude in our benchmarks (semgrep.dev)
   1112 points by jms703 8 days ago | 515 comments
2. Senior SWE-Bench: open-source benchmark that assesses agents as senior engineers (senior-swe-bench.snorkel.ai)
   184 points by matt_d 5 days ago | 116 comments
3. GLM 5.2 Performance Benchmarks (artificialanalysis.ai)
   164 points by theanonymousone 20 days ago | 48 comments
4. Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks (twitter.com)
   142 points by lucasfcosta 22 days ago | 45 comments
5. PostgresBench: A Reproducible Benchmark for Postgres Services (clickhouse.com)
   113 points by saisrirampur 16 days ago | 24 comments
6. Lies, Damn Lies and Database Benchmarks (questdb.com)
   58 points by eigenBasis 14 days ago | 28 comments
7. Claude Sonnet 5 – benchmark results (artificialanalysis.ai)
   41 points by lucamark 6 days ago | 17 comments
8. Why Weibo's tiny VibeThinker-3B has the AI world arguing over benchmarks again (venturebeat.com)
   19 points by gmays 18 days ago | 3 comments
9. Fable 5 remotion video benchmark and examples (mesmer.tools)
   9 points by mesmertech 27 days ago | 1 comments
10. Benchmark object storage in objects/s, not GB/s (fractalbits.com)
    8 points by zzsheng 13 days ago | discuss
11. MistralAI's Le Chaton Fat Tops Web Dev Benchmark
    7 points by nsoonhui 21 days ago | 1 comments
12. Ask HN: Are there good security benchmarks for LLMs?
    7 points by melvinroest about 12 hours ago | discuss
13. About Latency-focused disk benchmarks for Linux VPS environments (github.com)
    6 points by ashitlerferad 8 days ago | 1 comments
14. Every AI Memory Benchmark Has an Asterisk (tenureai.dev)
    6 points by freewilly25 12 days ago | discuss
15. DOJ Requires Egg Producers to End Coordinated Benchmark Manipulation (www.justice.gov)
    5 points by gnabgib 6 days ago | 1 comments
16. Can Java Microservices Be as Fast as Go? A 2026 Benchmark Update (medium.com)
    5 points by theanonymousone 19 days ago | 1 comments
17. A benchmark revealing an average memory-retrieval accuracy of 9% (zendoric.com)
    5 points by jflynt76 2 days ago | discuss
18. GPT-5.5-Cyber Tops Mythos 5 on Cybersecurity Benchmark (twitter.com)
    5 points by delichon 14 days ago | discuss
19. LLM benchmarks are answering someone else's question (danlevy.net)
    5 points by justsml 19 days ago | discuss
20. Show HN: We're inviting Anthropic to put the real Mythos 5 on our open benchmark (realvuln.com)
    4 points by jfaganel99 24 days ago | 4 comments
21. Chrome had the highest idle CPU and RAM usage in a 5-browser benchmark (gopeek-lovat.vercel.app)
    4 points by GeorgeWoff25 22 days ago | 3 comments
22. Show HN: TreasuryBench – an open benchmark for personal-finance AI advice (github.com)
    4 points by juneadkhan 11 days ago | 1 comments
23. How can Rust be so fast in the TechEmpower Web Framework Benchmarks? (kerkour.com)
    4 points by Keyb0ardWarri0r about 10 hours ago | discuss
24. General LLMs outperform specialized clinical AI tools on medical benchmarks (www.nature.com)
    4 points by sebg 11 days ago | discuss
25. One Line x86 Change to GCC Nets +12% Benchmark Win for Modern CPUs (www.phoronix.com)
    4 points by theanonymousone 12 days ago | discuss
26. BenchPress: Predict any LLM's score on any benchmark (microsoft.github.io)
    4 points by berlianta 13 days ago | discuss
27. DeepSWE Benchmark updated with GLM 5.2 and updated results for other models (deepswe.datacurve.ai)
    4 points by theanonymousone 16 days ago | discuss
28. General purpose LLMs outperform specialized clinical AI on medical benchmarks (www.nature.com)
    4 points by dnw 24 days ago | discuss
29. Show HN: Is grep enough? A transparent benchmark for agentic code navigation (entelligentsia.github.io)
    3 points by bonigv 7 days ago | 2 comments
30. DGX Spark Local LLM Benchmark: Administrative Tasks (www.aai-labs.com)
    3 points by tasubotadas about 4 hours ago | 1 comments