Gluon: Explicit Performance
(news.ycombinator.com)
31.
32.
33.
Nvidia's next AI chip may move beyond the all-purpose GPU
(techspot.com)
34.
Nvidia greenboost: transparently extend GPU VRAM using system RAM/NVMe
(news.ycombinator.com)
35.
36.
37.
38.
How to watch Jensen Huang’s Nvidia GTC 2026 keynote
(techcrunch.com)
39.
Samsung’s new tool wants to fix Android gaming’s biggest headache
(androidauthority.com)
40.
Launch HN: IonRouter (YC W26) – High-throughput, low-cost inference
(news.ycombinator.com)
41.
Show HN: Calyx – Ghostty-Based macOS Terminal with Liquid Glass UI
(news.ycombinator.com)
42.
BitNet: Inference framework for 1-bit LLMs
(news.ycombinator.com)
43.
Microsoft BitNet: 100B Param 1-Bit model for local CPUs
(news.ycombinator.com)
44.
AutoKernel: Autoresearch for GPU Kernels
(news.ycombinator.com)
45.
46.
FFmpeg-over-IP – Connect to remote FFmpeg servers
(news.ycombinator.com)
47.
Show HN: How I topped the HuggingFace open LLM leaderboard on two gaming GPUs
(news.ycombinator.com)
48.
Show HN: How I Topped the HuggingFace Open LLM Leaderboard on Two Gaming GPUs
(news.ycombinator.com)
49.
51.
52.
Building a Procedural Hex Map with Wave Function Collapse
(news.ycombinator.com)
53.
The 1979 Design Choice Breaking AI Workloads
(news.ycombinator.com)
54.
55.
Nvidia backs AI data center startup Nscale as it hits $14.6B valuation
(news.ycombinator.com)
56.
I ported Linux to the PS5 and turned it into a Steam Machine
(news.ycombinator.com)
57.
Show HN: Eyot, A programming language where the GPU is just another thread
(news.ycombinator.com)
58.
Autoresearch: Agents researching on single-GPU nanochat training automatically
(news.ycombinator.com)
59.
Helix: A post-modern text editor
(news.ycombinator.com)
60.
A tool that removes censorship from open-weight LLMs
(news.ycombinator.com)