llm on Hacker News

454 titles from 400 authors in the last 30 days, with 427 distinct links and 94 Show HN.

Counts titles, it does not confirm facts. Open the source before concluding anything.

Projects on this term

The score comes from our model.

• Show HN: Resin, fast native scheduler for LLM inference engines · score 7.8

  Resin provides a Rust-based scheduler and KV cache for LLM inference engines, separating scheduling logic from model-specific forward pass kernels in Python. It supports advanced features like speculative decoding, parallelisms, CUDA graphs, and an HTTP serve layer, allowing rapid development of custom inference engines with competitive performance.

• Show HN: Agentfork – Replay an LLM agent from any step and test fixes · score 7.8

  Agentfork is a Python tool for recording, replaying, and forking LLM agent runs at any step. It lets you sweep multiple prompt/model/tool variants with repeats to get pass rates, and save runs as CI tests. That turns flaky debugging anecdotes into measurable experiments with cost estimates, all locally.

• Show HN: LLM verification pipeline against hallucinations · score 7.8

  VeriQuote is a TypeScript library and CLI that enforces LLM citation verification through a two-stage pipeline: a deterministic fuzzy matcher confirms quotes exist in sources, and an LLM judge assesses whether quotes support claims. It addresses the critical problem of hallucinated citations in retrieval-augmented generation, with extensive benchmarks showing complementary blind spots between the two checks.

• Show HN: Open-source app that lets LLMs control any iPhone app (no Mac required) · score 7.5

  iphone-use-app is an open-source iOS application that lets an LLM control the same iPhone it runs on, using a loopback VPN and developer services without requiring a Mac at runtime or jailbreak. It is technically impressive, combining Swift and Rust to mount developer images and relay screenshots and tool calls, enabling autonomous app navigation. Its utility is currently limited by stringent requirements like iOS 26, Developer Mode, and a paid Apple developer account.

Recent titles

1. Uniichat: Use a single LLM chat forever
   2026-11-01 · github.com · HN discussion
2. Show HN: Modeljury – a jury of cheap LLMs that escalates the hard cases
   2026-11-01 · github.com · HN discussion
3. QwenPU: An LLM running inside another LLM's weights
   2026-11-01 · huggingface.co · HN discussion
4. Show HN: Rust crawler for SEO and content audits, with optional LLM review
   2026-11-01 · crawler.siteone.io · HN discussion
5. FineDNA: Open data to train LLMs on DNA
   2026-11-01 · huggingface.co · HN discussion
6. Open data and code to teach LLMs to read DNA
   2026-11-01 · github.com · HN discussion
7. Benchma(R)king: Making Chairs with LLMs
   2026-10-31 · emergencemagazine.org · HN discussion
8. Show HN: Hey Spark – open-source voice assistant for your laptop, any LLM brain
   2026-10-31 · github.com · HN discussion
9. The LLM Engineer's Schedule
   2026-10-31 · kwojcicki.github.io · HN discussion
10. Show HN: Reducing LLM Context Complexity from O(N^2) to O(k)
    2026-10-31 · github.com · HN discussion
11. GPUCost – Cloud GPU Prices Normalized by VRAM and LLM VRAM Sizer
    2026-10-31 · gpu-cost-r34p.vercel.app · HN discussion
12. UlaCache: Unified Logical Addressing for Sparse LLM Serving
    2026-10-31 · arxiv.org · HN discussion
13. PostgreSQL 19 Running Behind Schedule Due to AI/LLM Burden
    2026-10-31 · phoronix.com · HN discussion
14. Local Inference with LLMs
    2026-10-31 · jacquesmattheij.com · HN discussion
15. GAAC: Lossless KV Cache Compression for Faster LLM Inference
    2026-10-31 · arxiv.org · HN discussion
16. Gandalf: Extending SSD Lifespan for LLM KV Cache Offloading
    2026-10-31 · arxiv.org · HN discussion
17. Malware uses LLMs to rank victims by how much they're worth
    2026-10-31 · blog.talosintelligence.com · HN discussion
18. ChemEZ: LLM Wrapper with Chemistry Processing and Visualization
    2026-10-31 · github.com · HN discussion
19. Show HN: Tadpole – an LLM agent built from scratch then post-trained on 4B model
    2026-10-31 · github.com · HN discussion
20. Show HN: Free OCR Cost Calculator – Estimate multi-provider LLM token costs
    2026-10-31 · ocrbench.app · HN discussion
21. Show HN: Uncracked City, detective game where you question LLM suspects that lie
    2026-10-31 · uncracked.city · HN discussion
22. Real-time zero-knowledge LLM identification from physical telemetry
    2026-10-31 · zenodo.org · HN discussion
23. Non-autoregressive NC LLM spits tokens 21x faster than text diffusion
    2026-10-31 · hazyresearch.stanford.edu · HN discussion
24. Where AI agents' social boundaries break: 5 LLMs try to form a government
    2026-10-31 · judgementlabs.app · HN discussion
25. I built a honeypot for LLMs, will a agent take the bait?
    2026-10-30 · victorribeiro.com · HN discussion
26. Web-MCP-Bridge – Connect browser-based LLMs to MCP servers and tools
    2026-10-30 · github.com · HN discussion
27. Signet: Cryptographic Enforcement of LLM Transitions
    2026-10-30 · github.com · HN discussion
28. Show HN: NKTg AI – An LLM that decodes text before generating
    2026-10-30 · nktg.org · HN discussion
29. Learn LLM inference system on Apple Silicon
    2026-10-30 · github.com · HN discussion
30. Show HN: Anime Text Datasets for LLMs Training
    2026-10-30 · github.com · HN discussion

All trends