Homebench – Benchmark local LLMs for speed, memory, and quality
Article URL: https://github.com/david-g-3654/homebench Comments URL: https://news.ycombinator.com/item?id=49166308 Points: 4 # Comments: 0
Latest
The US is not a democracy but an oligarchy, study concludes (2014)
Article URL: https://www.upi.com/Top_News/US/2014/04/16/The-US-is-not-a-democracy-but-an-oligarchy-study-concludes/2761397680051/ Comments URL: https://news.ycombinator.com/item?id=49166254 Points: 14 # Comments: 4
An Honest Review of AI Programming
Article URL: https://mropert.github.io/2026/08/04/an_honest_review_of_ai_programming/ Comments URL: https://news.ycombinator.com/item?id=49166230 Points: 12 # Comments: 9
Time to finally migrate from LM Studio -> llama.cpp, your experience?
Has anyone moved from LM Studio to llama.cpp? What was your experience like? What did you have to learn in order to recreate your experience? Which…
FFmpeg 9.0
Article URL: https://github.com/FFmpeg/FFmpeg/blob/n9.0/RELEASE_NOTES Comments URL: https://news.ycombinator.com/item?id=49166202 Points: 34 # Comments: 1
Bending Spoons makes first post-IPO acquisition with $1.3B Airtable deal
Article URL: https://live.euronext.com/en/financial-news/bending-spoons-makes-first-post-ipo-acquisition-13-billion-airtable-deal Comments URL: https://news.ycombinator.com/item?id=49166182 Points: 6 # Comments: 1
b10258
llama : move n_vocab from llama_sampler_data to penalty_sampler (#26520) This matches how it is done for logit_bias and mirostat samplers, see #25262 (comment) Website: https://llama.app macOS/iOS:…
New DSv4 Flash Doom Loop in Q8? Llama.cpp Vulkan
Based on what everybody has been saying about this, I feel like I must've done something wrong. It was doing like 2 or 3 "Need maybe"…
The Downsides of LLM-Generated Peer Reviews [D]
Having used LLMs to assist with reviews, and also having received reviews that appear to rely heavily on LLM-generated text, I have noticed two recurring problems.…
DreamTraj: Generating 6-DoF Object Trajectories by Reading Unrendered Video Diffusion Latents
Accurate prediction of object trajectories during manipulation is essential for closing the perception-action loop. Progress is limited on two fronts: available datasets lack fine-grained language-to-motion annotations,…
24 Generative AI Adoption Statistics for Creative Teams in 2026
24 Generative AI Adoption Statistics for Creative Teams in 2026
GradCuit: Credit-Assigned Gradient Flow Enables Robust and Interpretable Test-Time Latent Reasoning
Optimization-based latent reasoning improves large language model outputs by optimizing instance-specific continuous states at test time while keeping model parameters frozen. Existing methods, however, typically connect…
GPTQ-2D: Cubic-Time Two-Sided Adaptive Rounding
Adaptive rounding methods such as GPTQ, or equivalently Babai's nearest plane algorithm, round a real matrix to integers under a quadratic metric. They process the entries…
Reflex Open Sources XY: A Rust-Backed Super-Fast Python Charting Library That Keeps 100 Million Point Charts Interactive
Reflex has released XY, an Apache-2.0 Python charting library that moves rendering work into a native Rust core and a WebGL2 client. It holds roughly 0.08…
Show HN: I made a private self-destructing image hosting site in Golang
So I'm a junior Go dev and I made a small site that allows me to share images with someone privately and securely, and give me…
RT PC Watch: Sakana AI、日本に特化したAIモデル「Sakana Namazu」提供開始 https://pc.watch.impress.co.jp/docs/news/2130476.html
RT PC WatchSakana AI、日本に特化したAIモデル「Sakana Namazu」提供開始 https://pc.watch.impress.co.jp/docs/news/2130476.html
Building an Advanced AI Skill Security Auditing Pipeline with NVIDIA SkillSpector, LangGraph, YARA Rules, SARIF, and CI Policy Gates
Learn how to build an end-to-end security assessment pipeline for AI agent skills using NVIDIA SkillSpector and LangGraph. In this tutorial, we construct a synthetic skill…
You don't need React: creating a minimal UI library in Vanilla JavaScript
Article URL: https://pedroth.github.io/?p=post/NoNeedReact Comments URL: https://news.ycombinator.com/item?id=49165665 Points: 19 # Comments: 5
Qwen3.8-Max, better and cheaper. Try it out. 👀
Qwen3.8-Max, better and cheaper. Try it out. 👀Command Code: We ran a test between the new Qwen3.8-Max, Opus 5 and GPT-5.6 Sol.3 models. same prompt. one-shot…
RT Sakana AI: Re Sakana Namazu：日本の感性で、考え、動くLLM Sakana Namazuは日本文化に対する深い理解と高性能な推論能力を兼ね備えたLLMです。Web検索やコ…
RT Sakana AIRe Sakana Namazu：日本の感性で、考え、動くLLMSakana Namazuは日本文化に対する深い理解と高性能な推論能力を兼ね備えたLLMです。Web検索やコード実行を組み合わせて様々なビジネスシーンにおける複雑なタスクをこなすことができます。今お使いのモデルから簡単に切り替えることができます。ぜひお試しください。https://sakana.ai/namazu/
Relax Within, Balance Across: Geometry-Guided Load Balancing for Vision-Language Mixture-of-Experts
Vision-language MoE batches contain different numbers of image and text tokens. Image resolution, image count, tiling, and prompt length all change this token mix. We call…
DeepVoyager-VL: Incentivizing Vision-in-the-Loop Search for Long-Horizon Multimodal Agents
Multimodal large language models (MLLMs) have advanced visual understanding and reasoning, yet their static parametric knowledge limits their ability to address knowledge-intensive and dynamically evolving open-world…
CollectWise (YC F24) Is Hiring
Article URL: https://www.ycombinator.com/companies/collectwise/jobs/oEAfzBS-ai-agent-engineer Comments URL: https://news.ycombinator.com/item?id=49165170 Points: 0 # Comments: 0
Motion Beyond Morphology: Bootstrapping Cross-Category Motion Transfer from Abstract Motion Representations
Video motion transfer aims to animate a target object using dynamics from a reference video. Existing formulations largely rely on fixed structural correspondence, which becomes ill-defined…