How people actually run agents: Ollama, local models, MCP servers, tooling threads.

Ollaya – Ollama for open-source, Jev-style decision models

ollaya.dev · 2026-09-25 618 pts 145 comments

From the creator of Redis; run LLM locally with ds4

dwarfstar.sh · 2026-10-02 367 pts 107 comments

Launch HN: Magnitude (YC S25) – Self-optimizing inference engine for agents

github.com · 2026-09-30 194 pts 98 comments

Show HN: JevBench, a reproducible benchmark for typed decision models

benchmarkheaven.com · 2026-09-22 154 pts 39 comments

Show HN: CUA-S1 – A System One Model for Computer Use

github.com · 2026-09-19 95 pts 10 comments

Faster prompt lookup drafting in llama.cpp

jadidbourbaki.github.io · 2026-09-26 89 pts 12 comments

Fourier Analysis: Drawing Llamas with Circles

adekau.github.io · 2026-09-24 79 pts 11 comments

Show HN: Rhun, an open-source code editor written in assembly

rhun.app · 2026-10-01 64 pts 51 comments

Show HN: Breadcrumb, record everything on your mac + context manager for AI

innerloop.works · 2026-10-01 49 pts 9 comments

Show HN: Apogee: Rebuilding Mozilla's Orbit, fully local and private

github.com · 2026-10-09 47 pts 2 comments

Show HN: Ekselio – Loveable for finance workflows (local first)

gptbeyond.com · 2026-09-25 43 pts 16 comments

Show HN: I created an open source locally usable full fledged AI platform

github.com · 2026-09-19 16 pts 13 comments

Show HN: Perspica – A semantic diff for reviewing code

github.com · 2026-09-30 15 pts 7 comments

ROCmFix and InferBench – AMD Local-LLM Setup and Vulkan vs. Hip Benchmarking

github.com · 2026-09-19 10 pts 2 comments

LLaMA Now Goes Faster on CPUs

justine.lol · 2026-10-02 6 pts 3 comments

New in Llama.cpp: Decision Models

huggingface.co · 2026-10-02 6 pts 1 comments

Show HN: Ollaya Web UI

github.com · 2026-09-26 6 pts 0 comments

Extract v2.5: Rebuilding extraction agents on LlamaParse

llamaindex.ai · 2026-10-01 5 pts 0 comments

Llama.cpp and WebGPU = Client-side LLMs [video]

youtube.com · 2026-10-07 4 pts 0 comments

Show HN: Flash-Agents – DSH as MCP for Claude

github.com · 2026-10-05 4 pts 0 comments

Ollama now supports Jev-like decision models all locally in 0.35

ollama.com · 2026-09-30 4 pts 0 comments

I got 2.2x more tokens per second from llama.cpp on Intel Arc

grigio.org · 2026-09-28 4 pts 0 comments

Llama.cpp Under the Hood

cppdepend.com · 2026-09-23 4 pts 0 comments

Transformers now runs llama.cpp quants

huggingface.co · 2026-09-23 4 pts 0 comments

Show HN: Nanosamur.ai – Ollama for Voice Models (STT)

github.com · 2026-09-21 4 pts 1 comments

Show HN: Kern Sandbox, run LLM-generated code locally, no cloud, no daemon

getkern.dev · 2026-09-28 4 pts 1 comments