cpu on Hacker News

64 titles from 52 authors in the last 30 days, with 53 distinct links and 9 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: FM-1 synth emulator with a reverse-engineered CPU and WebAssembly JIT · score 7.8

  A reverse-engineered emulator of the M-VAVE FM-1 pocket synthesizer, including its CPU, screen, keys, knobs, and audio, compiled to WebAssembly with a JIT so 81 community firmwares run unchanged in the browser, natively, and as AU/VST3 plugins. It enables firmware testing, development without hardware, and playing via MIDI, with persistent flash saves, backup export, and an embedded try-out API for third-party sites.

• Show HN: Enki LLM A pure rust (even kernels) with live GPU/CPU brain migration · score 7.0

  Enki LLM is a pure Rust implementation of Llama-3.2 1B inference that compiles Rust kernels directly to SPIR-V via the Enki runtime, eliminating CUDA dependencies and enabling live migration of model weights and KV-cache between GPU VRAM and CPU RAM mid-generation. The project demonstrates a novel heterogeneous compute approach with cross-silicon portability and zero-copy pointer-like memory access on the GPU.

• Show HN: CPU First Decision Model · score 7.0

  Gutsy is a fine-tuned 0.8B parameter model (based on Qwen3.5) that provides calibrated probabilities for decision-making questions instead of text generation. It runs efficiently on CPU via llama.cpp, offers a Jev-compatible inference API, and emphasizes privacy, determinism, and calibration for tasks like support triage, model routing, and fact checking.

• Show HN: Gutsy – small decision model for CPU (500MB), get probabilities back · score 7.0

  Gutsy is a small (0.8B parameter, CPU-runnable) fine-tuned decision model that returns calibrated probabilities for typed questions (yes/no, choice, score) about a given state, rather than generating text. It is technically notable for its probability-calibration training loss, state caching, order-robustness testing, and unusually candid weak-spot disclosures, while filling a niche gap between full LLMs and rule-based decision logic.

Recent titles

1. CPU Simulations Show Building Blocks of Life Didn't Arrive on Prehistoric Comets
   2026-11-01 · cnn.com · HN discussion
2. Making GROUP BY faster by doing more (CPU) work
   2026-10-30 · pivotlake.io · HN discussion
3. Qwenputer turns Gated DeltaNet state into exact RAM for a CPU
   2026-10-30 · theainewsreport.com · HN discussion
4. Parakeet v3 on a desktop CPU: speech to text at 20× real time
   2026-10-29 · talkhawk.app · HN discussion
5. Compile-Time Target CPU Feature Detection in C
   2026-10-29 · terroir.systems · HN discussion
6. Qwenputer: A functioning, deterministic RISC-V CPU inside Qwen3.5
   2026-10-29 · divagr.com · HN discussion
7. GitHub saw 7.4B commits in Sep., adding 4M CPU cores+160 PB storage to keep up
   2026-10-28 · thenewstack.io · HN discussion
8. A RISC-V CPU compiled into the weights of an LLM
   2026-10-28 · github.com · HN discussion
9. GPU vs. CPU. How to Cut Live Streaming and AI Processing Costs?
   2026-10-28 · red5.net · HN discussion
10. CPU Scheduling Explained [video]
    2026-10-27 · youtube.com · HN discussion
11. OpenAI's agents talked to each other through the CPU Cache
    2026-10-26 · the-decoder.com · HN discussion
12. Which Request Burned the CPU? Span-Tagged Profiles on Virtual Threads
    2026-10-26 · sku20.dev · HN discussion
13. Show HN: Arc-decide – an MIT 0.4B decision model for RAG checks that runs on CPU
    2026-10-26 · docs.docsgpt.cloud · HN discussion
14. AI agents, in separate VMs without network, communicated through the CPU cache
    2026-10-24 · twitter.com · HN discussion
15. A new way to fingerprint CPU architecture using Canvas and Skia
    2026-10-22 · scrapfly.dev · HN discussion
16. MicroLLM – Sub-millisecond calibrated System-1 decisions on CPU
    2026-10-22 · github.com · HN discussion
17. I built a game where you create a CPU from logic gates
    2026-10-22 · store.steampowered.com · HN discussion
18. Llmtobinary – LLM that generates bytes for a made up CPU archecture
    2026-10-22 · llmtobinary.com · HN discussion
19. Running ClickHouse on Nvidia Vera CPU
    2026-10-21 · clickhouse.com · HN discussion
20. I shipped an engine for your own UI: three DLLs, CPU, no CUDA required
    2026-10-20 · medium.com · HN discussion
21. Schematic, Mask, and Die Shot of Intel's 4004 CPU
    2026-10-19 · techav.net · HN discussion
22. Low-bit LLMs on CPU: what runs correctly and fast where
    2026-10-19 · github.com · HN discussion
23. Indra OCR: 1M-parameter OCR model that reads up to 49 pages per second on CPU
    2026-10-18 · github.com · HN discussion
24. Does moving a GPU server's request loop into the kernel save CPU?
    2026-10-18 · eunomia.dev · HN discussion
25. Hybrid ML-KEM costs bytes, not CPU: X25519 vs. ML-KEM key exchange benchmark
    2026-10-17 · github.com · HN discussion
26. Ryzen 9 9950X3D2 sold by Amazon was an Intel CPU under an AMD heat spreader
    2026-10-16 · reddit.com · HN discussion
27. Laptop Based on 6502 CPU
    2026-10-15 · maddox.pro · HN discussion
28. Show HN: BitRun v0.2 – Run 8B ternary LLMs on CPU with live wattage telemetry
    2026-10-15 · github.com · HN discussion
29. Gemma 3 1B at 114 tokens per second on a laptop CPU
    2026-10-15 · knl.co.in · HN discussion
30. Pareto Frontier Speech Recognition for CPU, Mobile and Edge
    2026-10-14 · futo.tech · HN discussion

All trends