A new runtime for k and q: l
Article URL: https://lv1.sh/ Comments URL: https://news.ycombinator.com/item?id=48821378 Points: 9 # Comments: 0
Communities 6900 stories
local already feels good enough
This is specifically for coding, technical planning, and hardware setup. _____ The only times Qwen 3.6 35B A3B has let me down, it has been something…
We're extending access to Fable 5 on all paid plans through July 12
https://xcancel.com/claudeai/status/2074548242386178258 Comments URL: https://news.ycombinator.com/item?id=48821102 Points: 59 # Comments: 44
Gepard : 0.6B streaming TTS built for real-time dialogue – 20× realtime factor, ~50ms time-to-first-audio, vLLM-native, Apache 2.0
We just open-sourced Gepard 1.0, a TTS model built for real-time conversation. It’s streaming-first: audio starts the moment text arrives, generated frame by frame instead of…
Mapping homes you can buy from the US government for <$100k
Article URL: https://govauctions.app/research/cheapest-homes-in-america Comments URL: https://news.ycombinator.com/item?id=48820274 Points: 10 # Comments: 5
I tested freshly merged DFlash in llama.cpp on Qwen 3.6 27B Local AI win. 4.44x faster at 36K context. Here are my findings RTX 6000 PRO.
Hey guys, A month ago I posted my MTP benchmarks here (3.34x on Gemma 4). DFlash support just merged into llama.cpp (PR #22105), so I ran…
Qwen3.6-27B – Effect of KV quantization on KLD – Q8, Q6, Q5 (bartowski)
Lower is better - Quantization increases from right to left I recently made a post here about how I squeezed more context into a Q8 model…
China sentences official to death for taking $325M in bribes
Article URL: https://www.bbc.com/news/articles/c33y0n1v1xjo Comments URL: https://news.ycombinator.com/item?id=48820220 Points: 13 # Comments: 11
Reducing Doom Loops with Final Token Preference Optimization
Article URL: https://www.liquid.ai/blog/antidoom Comments URL: https://news.ycombinator.com/item?id=48820127 Points: 3 # Comments: 0
Jim's TrueType QR Code Font
Article URL: https://github.com/jimparis/qr-font Comments URL: https://news.ycombinator.com/item?id=48820119 Points: 3 # Comments: 1
MacSurf 1.68 – NetSurf on OS 9 Released
Article URL: https://github.com/mplsllc/macsurf/releases/tag/v1.86 Comments URL: https://news.ycombinator.com/item?id=48820022 Points: 3 # Comments: 1
TorchJD: Training with multiple losses in PyTorch [P]
Hi everyone! I wanted to share some recent progress on TorchJD that might be useful to the machine learning community. When training models with multiple losses…
Weighing smoke: why AI visibility dashboards are mostly useless
Article URL: https://betterthangood.xyz/blog/weighing-smoke/ Comments URL: https://news.ycombinator.com/item?id=48819958 Points: 3 # Comments: 0
How are people hosting random GGUF / open models behind an API?
I keep running into this annoying gap: A model exists on Hugging Face. Sometimes it has GGUFs. It runs fine locally in Ollama / llama.cpp /…
Show HN: Rowboat – Open-source, local-first alternative to Claude Desktop
Claude’s desktop app is brilliant, but for our own daily work we kept wanting it to be less like a chat app and more like a…
30papers.com – Ilya's 30 essential ML papers, in a beginner friendly format
Article URL: https://30papers.com/ Comments URL: https://news.ycombinator.com/item?id=48819608 Points: 47 # Comments: 8
Sites that block AI training crawlers mostly ignore the answer time bots
Article URL: https://sitedex.dev/insights/robots-txt-2023-war-memorial Comments URL: https://news.ycombinator.com/item?id=48819517 Points: 10 # Comments: 0
Better Auth is joining Vercel
Article URL: https://better-auth.com/blog/better-auth-joins-vercel Comments URL: https://news.ycombinator.com/item?id=48819512 Points: 5 # Comments: 0
Why we built yet another Postgres connection pooler
Article URL: https://pgdog.dev/blog/why-yet-another-connection-pooler Comments URL: https://news.ycombinator.com/item?id=48819308 Points: 11 # Comments: 0
mistral.rs v0.9.0: up to 1.8x faster CPU decode than llama.cpp on x86 and ARM!
https://preview.redd.it/nuk5rxceptbh1.png?width=1448&format=png&auto=webp&s=300344dd4c6552379e8536b81ba288be3d6dca3f On Qwen3 4B Q4_K, mistral.rs decodes faster than llama.cpp at every context depth we measured, on x86 (Sapphire Rapids) and ARM (GB10). We optimized mistral.rs…
Software Bonkers
Article URL: https://craigmod.com/essays/software_bonkers/ Comments URL: https://news.ycombinator.com/item?id=48819250 Points: 8 # Comments: 3
Microsoft Fire IdTech Team at Id Software
Article URL: https://gamefromscratch.com/microsoft-fire-idtech-team-at-id-software/ Comments URL: https://news.ycombinator.com/item?id=48819244 Points: 45 # Comments: 6
Midtown Manhattan blocks evacuated after beams buckling at construction site
Article URL: https://abcnews.com/US/midtown-manhattan-blocks-evacuated-after-beams-found-buckling/story?id=134549272 Comments URL: https://news.ycombinator.com/item?id=48819232 Points: 12 # Comments: 7
Secret Tracker in Claude Code Uncovered, Anthropic Directly Deletes Code
Article URL: https://voi.id/en/technology/583552 Comments URL: https://news.ycombinator.com/item?id=48819084 Points: 7 # Comments: 1