
Distilling an LLM on One GPU: Offline Top-K Logits and a Fused Chunked KL Loss
LLM distillation on one GPU: cache the teacher's top-100 logits offline and use a fused chunked KL loss to cut a GPT-OSS 20B run from four nodes to one.
117 articles - page 3 of 10

LLM distillation on one GPU: cache the teacher's top-100 logits offline and use a fused chunked KL loss to cut a GPT-OSS 20B run from four nodes to one.

Muse Glimmer 30B is Meta's Apache 2.0 open-weight multimodal agent model that runs in 24GB. Benchmarks, hardware reality and what the numbers actually say.

A Claude-built night-sky site cloned an open source project's name, features and even a fixed bug. What it says about memorization and AI app verification.

WeatherNext Cyclones, DeepMind's open-source tropical cyclone model, adds a day of forecast lead time. What the Nature paper shows and how to run the weights.

OpenJDK's interim AI policy bans code generated in full or in part by LLMs. What it still allows, how it is enforced, and what it means for Java contributors.

Cloudflare OS is an open source agent workspace: capability-based Gatekeepers, apps as private per-user instances, and approvals that keep agents moving.

LFM2.5-2.6B is an open-weight on-device agent model from Liquid AI that runs in under 2.5 GB. How it was trained, what the benchmarks say and how to run it.

Prime Agent is Prime Intellect's open-source coding harness with one tool, a persistent IPython kernel. How it works, its benchmarks, and Claude Code compared.

Mistral Shieldstral is a 3B Apache 2.0 multimodal moderation model that takes your policy as a plain-language question at inference and runs on one 16GB GPU.

Vibe maintenance for AI-generated pull requests: Steve Yegge's workflow to automate triage, let agents repair good ideas, and keep maintainer taste in charge.

A ripgrep musl segfault in huge multi-threaded searches traced to a suspected Linux 7.0 kernel race. How it was pinned, plus the one-line fix candidate.

AI session portability: sealed reasoning, opaque compaction and hidden subagent messages are replacing portable transcripts. Five tests and seven rules for it.
Showing 25-36 of 116 articles

New tutorials, open-source projects, and deep dives on coding agents - delivered weekly.
Explore 840 topics
Browse All Topics