Hackernews posts about Opus 4.6

  1. Anonymous request-token comparisons from Opus 4.6 and Opus 4.7 (tokens.billchambers.me)
    615 points by anabranch 18 days ago | 576 comments
  2. Changes in the system prompt between Claude Opus 4.6 and 4.7 (simonwillison.net)
    370 points by pretext 18 days ago | 218 comments
  3. Claude Opus 4.6 accuracy on BridgeBench hallucination test drops from 83% to 68% (twitter.com)
    70 points by bratao 24 days ago | 20 comments
  4. GLM-5.1 matches Opus 4.6 in agentic performance, at ~1/3 actual cost (app.uniclaw.ai)
    22 points by skysniper 29 days ago | 2 comments
  5. Ask HN: Why Opus4.6 was silently removed from Claude Code?
    15 points by t0duf0du 15 days ago | 22 comments
  6. Enforcing new limits and retiring Opus 4.6 Fast from Copilot Pro+ (github.blog)
    15 points by ValentineC 26 days ago | 2 comments
  7. Who's on call? How Opus 4.6 helped us calculate this 2,500x faster (incident.io)
    11 points by boryrain 8 days ago | discuss
  8. Anthropic Nerfed Opus 4.6 Before the 4.7 Launch (fagnerbrack.com)
    8 points by fagnerbrack 19 days ago | 4 comments
  9. Is it just me, or Opus 4.6 is sounding bit dumb lately
    7 points by rambrrest 27 days ago | 5 comments
  10. Comparing GPT-5.4, Opus 4.6, GLM-5.1, Kimi K2.5, MiMo V2 Pro and MiniMax M2.7 (www.codejam.info)
    6 points by funkyval 18 days ago | 2 comments
  11. Ask HN: Models Comparable to Opus 4.6?
    5 points by islandbytes 13 days ago | 3 comments
  12. Changes in the system prompt between Claude Opus 4.6 and 4.7 (simonwillison.net)
    5 points by simonw 18 days ago | discuss
  13. GitHub Copilot Pro+ not allowing Claude Opus 4.6 (github.com)
    4 points by vikrantrathore 16 days ago | 1 comments
  14. Enforcing new limits and retiring Opus 4.6 Fast from Copilot Pro+ (github.blog)
    3 points by GaryBluto 23 days ago | 4 comments
  15. Opus 4.7 dominates agentic benchmark, 15% more expensive than Opus 4.6 (app.uniclaw.ai)
    3 points by skysniper 20 days ago | 1 comments
  16. Tell HN: Opus 4.6/4.7 cyber policy changes break authorized bug bounty workflows
    3 points by johnmlussier 20 days ago | discuss
  17. How good is Opus 4.6 at vuln detection? (zeropath.com)
    3 points by NonStopOyster 29 days ago | discuss
  18. Claude Opus 4.6 vs. Opus 4.7 Effort Levels and Prompt Steering Benchmarks (ai.georgeliu.com)
    2 points by mszel 10 days ago | 1 comments
  19. I ran Opus 4.7 vs. Old Opus 4.6 vs. New Opus 4.6 on 28 Zod tasks (www.stet.sh)
    2 points by bisonbear 19 days ago | discuss
  20. Opus 4.6 is getting BAD [video] (www.youtube.com)
    1 points by zadkey 23 days ago | 1 comments
  21. Claude Opus 4.7 vs. Claude Opus 4.6: What Changed? (www.mindstudio.ai)
    1 points by sminchev 5 days ago | discuss
  22. Claude Opus 4.6 was nerfed prior to release of Opus 4.7 (twitter.com)
    1 points by nomilk 13 days ago | discuss
  23. Ask HN: At ~165k tokens, does Opus 4.6 1M outperform Opus 4.6 200k?
    1 points by consumer451 22 days ago | discuss
  24. N-Day-Bench – Can LLMs find real vulnerabilities in real codebases? (ndaybench.winfunc.com)
    91 points by mufeedvh 23 days ago | 29 comments
  25. Show HN: Eve – Managed OpenClaw for work (eve.new)
    72 points by zachdive 26 days ago | 40 comments
  26. Launch HN: Relvy (YC F24) – On-call runbooks, automated (www.relvy.ai)
    48 points by behat 28 days ago | 25 comments
  27. Opus 4.7 is horrible at writing
    19 points by limalabs 20 days ago | 27 comments
  28. Opus 4.7 vs. 4.6 after 3 days of real coding side by side from my actual session
    16 points by agentseal 17 days ago | 4 comments
  29. Why Codex works better than Claude Code for my production monolith
    14 points by anophelon 8 days ago | 2 comments
  30. Show HN: I built a database for AI agents (github.com)
    12 points by Kappa90 29 days ago | 10 comments
More