Hackernews posts about ARC-AGI

  1. Tripling an LLM's ARC-AGI-2 score with code evolution (imbue.com)
    17 points by danielmewes 21 days ago | 2 comments
  2. I beat Grok 4 on ARC-AGI-2 using a CPU-only symbolic engine (18.1% score) (github.com)
    11 points by kofdai 24 days ago | 6 comments
  3. Show HN: Launch an AI agent swarm for ARC-AGI-3 with plain-English prompts (arc-agi-swarm.vercel.app)
    3 points by surferbayarea about 11 hours ago | 1 comments
  4. Saturated ARC-AGI-2 (www.ycombinator.com)
    3 points by eightnoteight 25 days ago | 1 comments
  5. Noemon: ARC-AGI-2 top score at a fraction of the cost (twitter.com)
    2 points by markerbrod 14 days ago | 1 comments
  6. Open Source Models Score Low on ARC-AGI-2 Reasoning Benchmark (xcancel.com)
    2 points by ironyman 18 days ago | discuss
  7. ARC-AGI-3 Update (twitter.com)
    2 points by gmays 29 days ago | discuss
  8. 44% on ARC-AGI-1 in 67 cents (github.com)
    1 points by evilmathkid 16 days ago | 1 comments
  9. Hill-climbing ARC-AGI-3 (blog.alexisfox.dev)
    1 points by 10xDev 11 days ago | discuss
  10. AI Broke ARC-AGI-2 at 84.6% – But the Key Trick Is from 1972 (ai.gopubby.com)
    1 points by Aedelon 21 days ago | discuss
  11. Pure LLMs Score 0% on ARC-AGI-2. Why the Third Wave of AI Looks Like the First (ai.gopubby.com)
    1 points by Aedelon 21 days ago | discuss
  12. Show HN: Optimize_anything: A Universal API for Optimizing Any Text Parameter (gepa-ai.github.io)
    8 points by LakshyAAAgrawal 29 days ago | discuss
  13. Show HN: We Built Kaggle for AI Agents (hive.rllm-project.com)
    7 points by deviscold 1 day ago | 2 comments
  14. Is anyone compressing AI models for the 4B people without GPUs or internet?
    3 points by yashpxl 19 days ago | 1 comments
  15. Ask HN: Are AI mandates a good idea?
    4 points by Solar_Flare about 3 hours ago | 9 comments
  16. Show HN: Ripl – A unified 2D/3D engine for Canvas, SVG, WebGPU, and the Terminal (www.ripl.rocks)
    17 points by andrewcourtice 1 day ago | discuss
  17. Show HN: Unwind – I built a security proxy for AI agents on a Raspberry Pi
    3 points by David-Brug-Ai 1 day ago | discuss
  18. Free market research reports covering every tech sector (github.com)
    1 points by aimarketintel about 21 hours ago | discuss
  19. Show HN: Pitstop-check – finds the retry bug that turns 429s into request storms
    1 points by SirBrenton about 13 hours ago | 1 comments
  20. OpenAI O3 breakthrough high score on ARC-AGI-PUB (arcprize.org)
    1724 points by maurycy over 1 year ago | 1755 comments
  21. Getting 50% (SoTA) on Arc-AGI with GPT-4o (redwoodresearch.substack.com)
    394 points by tomduncalf almost 2 years ago | 243 comments
  22. ARC-AGI without pretraining (iliao2345.github.io)
    351 points by georgehill about 1 year ago | 121 comments
  23. Arc-AGI-2 and ARC Prize 2025 (arcprize.org)
    188 points by gkamradt 12 months ago | 101 comments
  24. OpenAI o1 Results on ARC-AGI-Pub (arcprize.org)
    187 points by z7 over 1 year ago | 118 comments
  25. I got the highest score on ARC-AGI again swapping Python for English (jeremyberman.substack.com)
    178 points by freediver 6 months ago | 131 comments
  26. O3 "Arc AGI" Postmortem (garymarcus.substack.com)
    44 points by signa11 about 1 year ago | 16 comments
  27. Samsung released a 7M model that achieved 45% on ARC-AGI-1 (arxiv.org)
    34 points by chintler 5 months ago | 12 comments
  28. Hidden drivers of HRM's performance on ARC-AGI (arcprize.org)
    31 points by veryluckyxyz 5 months ago | 2 comments
  29. Poetiq shatters ARC-AGI 2 benchmark at half the cost (poetiq.ai)
    25 points by flavio87 4 months ago | 3 comments
  30. Grok 4 (Thinking) achieves new SOTA on ARC-AGI-2 (X) (twitter.com)
    17 points by mellosouls 8 months ago | discuss
  31. ARC-AGI 2: Poetiq reaches 75% at less $8 / task (twitter.com)
    13 points by dimtion 3 months ago | 1 comments
  32. Playable preview of ARC-AGI-3 (three.arcprize.org)
    11 points by dcre 8 months ago | 2 comments
  33. Arc‑AGI Without Pretraining (twitter.com)
    10 points by E-Reverance about 1 year ago | 1 comments
  34. Show HN: ARC-AGI-3 Toolkit (docs.arcprize.org)
    9 points by gkamradt about 2 months ago | 1 comments
  35. ARC-AGI-2 human baseline surpassed (www.lesswrong.com)
    9 points by hugetim 3 months ago | 1 comments
  36. The Drivers of HRM's Performance on Arc-AGI (arcprize.org)
    8 points by artninja1988 7 months ago | discuss
  37. Agentic systems redraw the Pareto frontier on ARC-AGI (poetiq.ai)
    7 points by gkapur 4 months ago | 1 comments
  38. Analyzing o3 and o4-mini with ARC-AGI (arcprize.org)
    6 points by artninja1988 11 months ago | 1 comments
  39. ARC-AGI 2025: A research review (lewish.io)
    6 points by kordlessagain 9 months ago | discuss
  40. New SOTA on Arc-AGI Using Grok 4 (twitter.com)
    5 points by Rover222 6 months ago | 3 comments
  41. ARC-AGI-3 Preview (three.arcprize.org)
    5 points by blixt 7 months ago | 2 comments
  42. [ARC-AGI-2 SoTA] Efficient Evolutionary Program Synthesis (ctpang.substack.com)
    5 points by georgehill 6 months ago | discuss
  43. F. Chollet: We were able to reproduce the findings of the HRM paper on ARC-AGI-1 (twitter.com)
    5 points by Anon84 7 months ago | discuss
  44. LLMs struggle with perception, not reasoning, in ARC-AGI (anokas.substack.com)
    5 points by s-macke about 1 year ago | discuss
  45. Gemini achieving "incredible numbers" (84.6%) on ARC-AGI-2 (Chollet) (twitter.com)
    4 points by marojejian about 1 month ago | 2 comments
  46. Arc AGI 2025 (arcprize.org)
    4 points by artninja1988 12 months ago | 2 comments
  47. Poetiq achieves 75% at under $8 / problem using GPT-5.2 X-High on ARC-AGI-2 (poetiq.ai)
    4 points by mromanuk 3 months ago | discuss
  48. Poetiq: SOTA Reasoning on ARC-AGI (github.com)
    4 points by gfortaine 4 months ago | discuss
  49. Visual Reasoning and Tool Use Double GPT-5's Arc-AGI-2 Success Rate (github.com)
    4 points by zoecarver 7 months ago | discuss