Ai/ml Infrastructure RSS Feeds

Directory0 ai/ml infrastructure feeds · 9 articlesShow all languages →

Latest

Ai/ml Infrastructure articles

Researchers Serve 229B-Parameter MoE Model Across Five Consumer GPUs Over Public Internet

We serve MiniMax-M2.5, a 229B-parameter mixture-of-experts model, split across five consumer RTX 5090s in five European countries. The stages share nothing but the public internet. Single-stream decoding reaches 10-13 tokens/s on interactive reasoning and

Insighttechnologyprogrammingdistributed systems

doi.org·Twitter AI (trending)·14d ago·2 min read

FLUX.2 [dev] API Provider Benchmark: Latency, Speed & Price Comparison

Insighttechnologyprogramming

Shared by @clattner_llvm ↗

artificialanalysis.ai15d ago

KVarN: Huawei's Calibration-Free vLLM KV-Cache Quantization Backend for Agentic Workloads

Codetechnologyprogramming

github.com1mo ago

Cost Analysis: Running Local LLMs on Apple Silicon vs. OpenRouter API

Insighttechnologyprogramming

williamangel.net2mo ago

Burla demo: Analyzing 1.7M Airbnb photos and 50.7M reviews at scale using CLIP and Claude Haiku Vision

119 cities x 4 quarterly snapshots of Inside Airbnb data. 1.7M photos through CLIP, the most suspicious shortlists double-checked with Claude Haiku Vision, 50.7M reviews scored through Haiku, all parallelized on Burla.

technologycloud computingdata scienceburla-cloud.github.io·Hacker News: Front Page·2mo ago·2 min read

Systalyze's Utilyze Tool Reveals True GPU Compute Utilization in AI Workloads

Systalyze's groundbreaking platform uncovers and eliminates inefficiencies in AI workloads, enabling customers to cut costs by up to 90%

Insighttechnologyprogramminggpu computing

systalyze.com·Hacker News: Front Page·2mo ago·19 min read

LMSYS Announces Day-0 Open-Source Support for DeepSeek-V4 with SGLang and Miles Stack

We are thrilled to announce Day-0 support for DeepSeek-V4 across both inference and RL training. SGLang and Miles form the first open-source stack to serve and train DeepSeek-V4 on launch day — with s...

Newstechnologyprogrammingopen sourcelmsys.org·Hacker News: Front Page·2mo ago·18 min read

Proposal for MCP Apps Extension (SEP-1865) to Standardize Interactive User Interfaces in Model Context Protocol

Newstechnologyprogramming

blog.modelcontextprotocol.io7mo ago

Fluid Storage: Forkable, Ephemeral Infrastructure for AI Agent Era

technologyinfrastructure

tigerdata.com8mo ago