Hackernews posts about Tolkien

Tolkien is a renowned English author, philologist, and professor best known for creating the fictional Middle-earth world and characters in his fantasy stories.

1. My security camera shipped a GitHub admin token in its login page (hhh.hn)
   642 points by hhh 9 days ago | 243 comments
2. GigaToken: ~1000x faster Language model tokenization (github.com)
   620 points by syrusakbary 11 days ago | 120 comments
3. GPT-5.5 Codex reasoning-token clustering may be leading to degraded performance (github.com)
   372 points by maille 29 days ago | 152 comments
4. Running Gemma 4 26B at 5 tokens/sec on a 13-year-old Xeon with no GPU (www.neomindlabs.com)
   329 points by neomindryan 18 days ago | 212 comments
5. The relay market powering token resellers and fraud (vectoral.com)
   207 points by mlenhard 7 days ago | 136 comments
6. I burned all my tokens researching how to save tokens (quesma.com)
   181 points by bkotrys 14 days ago | 213 comments
7. Price per 1M tokens is meaningless (janilowski.pl)
   142 points by janilowski 27 days ago | 103 comments
8. TokenTown: A visual way to understand how LLMs work (laurentiugabriel.github.io)
   73 points by laurentiurad 4 days ago | 21 comments
9. Launch HN: Tokenless (YC S26) – Automatic model switching to save money (usetokenless.com)
   71 points by rohaga 4 days ago | 63 comments
10. Show HN: Docx-CLI: agents read/edit Word docs using 1/2 the time and tokens (github.com)
    70 points by kirillklimuk 26 days ago | 31 comments
11. Brazilian farmers tokenized dairy cows to get loans, bypassing bank limits (www.coindesk.com)
    61 points by paulpauper 8 days ago | 54 comments
12. Ways to think about token pricing (www.ben-evans.com)
    46 points by mercutio2 24 days ago | 22 comments
13. Does Speaking to Agents Like Cavemen Save 65% of Tokens? We Test (blog.jetbrains.com)
    45 points by Sandman 5 days ago | 41 comments
14. Reducing Doom Loops with Final Token Preference Optimization (www.liquid.ai)
    32 points by dataminer 26 days ago | 10 comments
15. Unlimited AI tokens aren't unlimited after all as US Army burns through supply (arstechnica.com)
    25 points by Bender 11 days ago | 7 comments
16. Not everything should cost a token: the case for deterministic AI (www.vybe.build)
    24 points by marwann 27 days ago | 18 comments
17. $85,000 in tokens later: What I learned from scaling agentic coding at Lovable (lovable.dev)
    21 points by aliclark 29 days ago | 55 comments
18. SigMap: 97% token reduction for AI coding sessions (sigmap.io)
    18 points by handfuloflight 28 days ago | 5 comments
19. I sent Claude Opus 5 '–-' and it wrote me 5k tokens about a cartographer (austinsnerdythings.com)
    16 points by auspiv 5 days ago | 6 comments
20. Palo Alto CEO Arora says AI pricing needs to fall 90% as token costs skyrocket (www.cnbc.com)
    13 points by petethomas 24 days ago | 2 comments
21. Show HN: An unmetered LLM API–$6/month, no token tracking, no limits (yolo-auto.com)
    12 points by yolo-auto 28 days ago | 13 comments
22. Save Claude Code Tokens with Smart Routing (github.com)
    11 points by FrancescoMassa 30 days ago | 3 comments
23. Show HN: Awsmux – Multi-account AWS CLI, up to 5.4x faster, 7.4x fewer tokens (github.com)
    10 points by 0hardik1 8 days ago | 4 comments
24. Headroom - compress AI agent input for reduced token usage w/out harming output (github.com)
    10 points by andsoitis 12 days ago | 2 comments
25. Nvidia Vera Rubin Driving Performance per Watt, Lowest Token Cost for Partners (www.hpcwire.com)
    10 points by rbanffy 12 days ago | 2 comments
26. Token prices collapsing, regulation rising, AI's pricing power looks fragile (www.latimes.com)
    10 points by 1vuio0pswjnm7 30 days ago | discuss
27. 40–90% fewer tokens on Claude Code via TokenOptimization (github.com)
    9 points by brianiterate 12 days ago | discuss
28. Show HN: A 6M-token movable window on a single 46GB GPU (arxiv.org)
    7 points by Wetime 6 days ago | 16 comments
29. Wattage: A token-spend profiler and cost-regression gate for AI agents (github.com)
    6 points by faizanraza03 7 days ago | 2 comments
30. DeepSeek V4 Is Earning Agentic Token Share (openrouter.ai)
    6 points by verdverm 27 days ago | 2 comments