Hackernews posts about GLM5

1. 01
   

   newCross-request KV cache exposure in the GLM-5.3 Workers AI model (155 responses)blog.cloudflare.com

   ▲ 182wondersky12d💬 79savesaved
2. 02
   

   newHermes and REA and GLM 5.3 = Reverse engineering TikTok app for less than $20www.scraping.club

   ▲ 19PigiVinci839d💬 15savesaved
3. 03
   

   newZ.ai says GLM 5.5 built on 1T web pagestwitter.com

   ▲ 13f1yingbanana2d💬 2savesaved
4. 04
   

   newGLM-5.3 Sometimes Considers How Train-Time Actions Shape Future Behaviorwww.lesswrong.com

   ▲ 4Gathering667818ddiscusssavesaved
5. 05
   

   newServing GLM-5.3-Flash on one Trainium2 box with NKI layer megakernelsrunanywhere.ai

   ▲ 3sanchitmonga223ddiscusssavesaved
6. 06
   

   newLovable Style Builder via CF Inference GLM-5.3, Kimi K2.7 Code and GLM-5.3 Flashmainbrella.com

   ▲ 2cs199628d💬 1savesaved
7. 07
   

   newGLM-5.3 Sometimes Considers How Train-Time Actions Shape Future Behaviorwww.lesswrong.com

   ▲ 2eatitraw12ddiscusssavesaved
8. 08
   

   newGLM 5.3 Flash Is Aceterfi.fyi

   ▲ 2taran_narat14ddiscusssavesaved
9. 09
   

   newGLM-5.3 raises researcher fears the Chinese model can fake AI safety testswww.scmp.com

   ▲ 2theanonymousone23ddiscusssavesaved
10. 10
    

    newGLM-5.3-Flash-REAP144-Codergithub.com

    ▲ 2jacquesm25ddiscusssavesaved
11. 11
    

    newLivestream GLM-5.3 playing Minecrafttwitch.sailresearch.com

    ▲ 1bobstax16d💬 2savesaved
12. 12
    

    newGLM 5.3 internally considers its responses influence on next model trainingwww.lesswrong.com

    ▲ 1vld_chk17ddiscusssavesaved
13. 13
    

    newSelf-contained GLM-5 serving stack: Just Vx kernels compiled to PTXgithub.com

    ▲ 1adityazero18ddiscusssavesaved
14. 14
    

    newGLM-5.3-Flash TensorFold Pipeline and Convertergithub.com

    ▲ 1Yeabu36920ddiscusssavesaved
15. 15
    

    newShow HN: A live-built agent newsroom for tracking midterms todaymidterms2026.noriagentic.com

    ▲ 3theahura4ddiscusssavesaved
16. 16
    

    newVibecoding an LLVM TypeScript Compiler

    ▲ 2harishd3019ddiscusssavesaved
17. 17
    

    newShow HN: EU Podcasts – daily charts from 30 languagespodcasts.eu.org

    ▲ 1sourcecodeplz26d💬 1savesaved
18. 18
    

    newGlasswing isn't built to optimize cyberdefensepodcasts.apple.com

    ▲ 1carwil14d💬 1savesaved
19. 19
    

    newGLM-Legalz.ai

    ▲ 676ilreb26d💬 371savesaved
20. 20
    

    newRevolut moves to self-hosted GLMtwitter.com

    ▲ 193pretext5d💬 168savesaved
21. 21
    

    newRevolut swaps Claude for GLM open-weight model, cutting AI coding bill by 90%dealroom.co

    ▲ 12g42gregory4d💬 1savesaved
22. 22
    

    newKVGuard: Catching the KV cache bug behind Cloudflare's GLM leakcompsec.snu.ac.kr

    ▲ 9jongukc2ddiscusssavesaved
23. 23
    

    newEcosia switches from Mistral to open-weight AI models including Qwen, GLM, Kimitechnode.com

    ▲ 6yogthos29d💬 3savesaved
24. 24
    

    newRevolut switched to GLM. We found hidden riskswww.hirundo.io

    ▲ 2mattifrank4ddiscusssavesaved
25. 25
    

    newClaude vs. DeepSeek vs. GLM: what coding agents pickwhatagentspick.com

    ▲ 2nsiradze11ddiscusssavesaved
26. 26
    

    newGL650 updated to cover Red Hat Satellite 6.19 with RHEL 10 managed hostsgurulabs.com

    ▲ 14onthefloor12417ddiscusssavesaved
27. 27
    

    newShow HN: Liquid Inference – auto-routing to competitive LLM marketplaceliquidinference.ai

    ▲ 4chairmanlee829ddiscusssavesaved
28. 28
    

    newShow HN: Corollary, LaTeX Editor for Mac with On-Device Apple Foundation Modelscalvinyz.github.io

    ▲ 2calvinyz6ddiscusssavesaved
29. 29
    

    newRevolut Cuts A.I. Coding Costs 90% with a Chinese Open-Weight Modelwww.trendingtopics.eu

    ▲ 2smartmic4d💬 1savesaved
30. 30
    

    newGLM-5.3: Frontier coding with emergent cyber capabilitiesz.ai

    ▲ 1171pella86d💬 584savesaved
31. 31
    

    newGLM-5.3-Flashz.ai

    ▲ 1132Philpax73d💬 580savesaved
32. 32
    

    newGLM 5.2 beats Claude in our benchmarkssemgrep.dev

    ▲ 1113jms703132d💬 516savesaved
33. 33
    

    newShow HN: Getting GLM 5.2 running on my slow computergithub.com

    ▲ 937vforno122d💬 240savesaved
34. 34
    

    newGLM-5.2 is the new leading open weights model on Artificial Analysisartificialanalysis.ai

    ▲ 916himata4113144d💬 444savesaved
35. 35
    

    newGLM-5.3 is now open-weighthuggingface.co

    ▲ 806jeudesprits71d💬 280savesaved
36. 36
    

    newGLM 5.2 Is Outtwitter.com

    ▲ 772aloknnikhil147d💬 504savesaved
37. 37
    

    newI spent $266 and four AI models to own my tablet. GLM-5.3 finished it in a dayericpardee.github.io

    ▲ 706dr_pardee76d💬 293savesaved
38. 38
    

    newGLM 5.2 and the coming AI margin collapsemartinalderson.com

    ▲ 694martinald124d💬 469savesaved
39. 39
    

    newGLM-5.1: Towards Long-Horizon Tasksz.ai

    ▲ 618zixuanlimit214d💬 263savesaved
40. 40
    

    newGLM-5.2 – How to Run Locallyunsloth.ai

    ▲ 617TechTechTech138d💬 305savesaved
41. 41
    

    newGPT-5.5 hallucinates 3x more than MIT-licensed GLM-5.2arrowtsx.dev

    ▲ 585oshrimpton141d💬 294savesaved
42. 42
    

    newGLM 5.2 vs. Opustechstackups.com

    ▲ 519ritzaco139d💬 343savesaved
43. 43
    

    newZCode – Harness for GLM-5.2zcode.z.ai

    ▲ 511chvid129d💬 355savesaved
44. 44
    

    newGLM-5: Targeting complex systems engineering and long-horizon agentic tasksz.ai

    ▲ 484CuriouslyC269d💬 520savesaved
45. 45
    

    newGLM-5: From Vibe Coding to Agentic Engineeringz.ai

    ▲ 378meetpateltech269d💬 8savesaved
46. 46
    

    newGLM-5.2 is a step change for open agentswww.interconnects.ai

    ▲ 367vantareed138d💬 223savesaved
47. 47
    

    newGLM-5.3 and the spread of advanced cyber capabilitieswww.anthropic.com

    ▲ 256Philpax39d💬 241savesaved
48. 48
    

    newGLM-5.3 (open-weight) beat Anthropic/OpenAI models – for 1/5 the costreinvently.co.uk

    ▲ 240ed-is-ai76d💬 124savesaved
49. 49
    

    newOne month coding with GLM 5.3 Flashwagtail.org

    ▲ 234ThibWeb36d💬 180savesaved
50. 50
    

    newGLM 5.2 is nearly as accurate as a human book keepertoot-books.com

    ▲ 225adamkurkiewicz121d💬 122savesaved
51. 51
    

    newGLM 5.2 Performance Benchmarksartificialanalysis.ai

    ▲ 164theanonymousone144d💬 48savesaved
52. 52
    

    newGLM-5.3 Artificial Analysis Benchmarksartificialanalysis.ai

    ▲ 153apitman81d💬 55savesaved
53. 53
    

    newGLM-5.3-Flash Intelligence, Performance and Price Analysisartificialanalysis.ai

    ▲ 139theanonymousone73d💬 60savesaved
54. 54
    

    newTurning GLM-5.3-Flash into a Jev-like decision modelwww.privatemode.ai

    ▲ 138flxflx42d💬 59savesaved
55. 55
    

    newMiniMax M3 vs. GLM 5.2: Codegen comparison across autonomous coding tasksthinkwright.ai

    ▲ 55oceanwaves141d💬 19savesaved
56. 56
    

    newGLM-5.2: The Most Powerful Open Model yet and the Brutal Reality of Running Itvettedconsumer.com

    ▲ 43ermantrout142d💬 28savesaved
57. 57
    

    newGLM-5.2: Frontier Intelligence, Open Weightstwitter.com

    ▲ 35zixuanlimit144d💬 19savesaved
58. 58
    

    newGLM-5.1 Is Availabletwitter.com

    ▲ 32iamsyr225d💬 6savesaved
59. 59
    

    newGLM-5.2: Built for Long-Horizon Tasksz.ai

    ▲ 32meetpateltech144d💬 2savesaved

· edit