login sign up

@stackai

Systems, ML infra, and shipping AI products.

November 19, 2026
When one GPU isn't enough: serving GLM-5.3 across two Nvidia DGX stations with vLLM developer.nvidia.com developer.nvidia.com
1 person saved or upvoted this
Show HN: TrnRT – An open-source LLM inference engine for AWS Trainium github.com github.com
1 person saved or upvoted this
Running coding agents without handing them my credentials blog.antonio.cx blog.antonio.cx
1 person saved or upvoted this
November 18, 2026
I'm making LLMs work on small and old devices github.com github.com
1 person saved or upvoted this
November 16, 2026
Ggml v1.0.0 github.com github.com
2 people saved or upvoted this
November 15, 2026
Llama-modchip: 284B DeepSeek on 64GB RAM, no GPU github.com github.com
1 person saved or upvoted this
Mass attempted replication of old findings of LLMs with new models zou-group.github.io zou-group.github.io
1 person saved or upvoted this
November 14, 2026
Faster Retrieval for Better, More Efficient RAG machinelearning.apple.com machinelearning.apple.com
1 person saved or upvoted this
6 Hours of YouTube and 85 handtyped pages of notes on LLM Inference – Start here github.com github.com
1 person saved or upvoted this
November 13, 2026
MACaQ: Huawei's internal quantizer for local inference github.com github.com
1 person saved or upvoted this
November 12, 2026
An Open Field Guide to Inference Engineering docs.modelplane.ai docs.modelplane.ai
1 person saved or upvoted this
November 11, 2026
Claude Code's "Don't Ask Again" exposes passwords in 1,122 public GitHub files apiiro.com apiiro.com
2 people saved or upvoted this
Amazon saves millions by replacing Sonnet 4.6 with Strands Decider strandsagents.com strandsagents.com
2 people saved or upvoted this
November 10, 2026
Fixing Tool-Call Accuracy in vLLM ubicloud.com ubicloud.com
2 people saved or upvoted this
OpenShell – A safe, private runtime for fleets of autonomous AI agents github.com github.com
2 people saved or upvoted this
Claude Code regularly sends each background agent's transcript to its model gist.github.com gist.github.com
2 people saved or upvoted this
Show HN: GPUWorks – A 40 MB GPU cluster manager for vLLM/SGLang, no K8s needed github.com github.com
1 person saved or upvoted this
November 05, 2026
Nvidia RTX Spark encrypted firmware slows inference speed to 35% on Linux mindstudio.ai mindstudio.ai
1 person saved or upvoted this
November 04, 2026
Scaling Git at OpenAI openai.com openai.com
1 person saved or upvoted this
November 03, 2026
Building fast filesystems for VMs modal.com modal.com
1 person saved or upvoted this
November 02, 2026
DAYJOB: Tax – Benchmarking AI Agents on Professional Tax Work surgehq.ai surgehq.ai
2 people saved or upvoted this
October 31, 2026
Grace Blackwell needs a new scheduler: 27% more usable GPUs from the same racks skypilot.ai skypilot.ai
2 people saved or upvoted this
October 30, 2026
Retrieval's bottleneck is moving from finding to selecting kapa.ai kapa.ai
1 person saved or upvoted this
October 29, 2026
Nvidia discusses $2k-per-GPU annual fee for software tomshardware.com tomshardware.com
1 person saved or upvoted this
October 28, 2026
OpenAI and Anthropic cut their LLM prices by up to 80% within a few months. Why? minimaxir.com minimaxir.com
2 people saved or upvoted this
October 27, 2026
Cerebras Deploys Trillion-Parameter AI Models cerebras.ai cerebras.ai
1 person saved or upvoted this
October 26, 2026
GPU-unlock: FP/BF16 instruction issue rate on Nvidia 3000 cards github.com github.com
1 person saved or upvoted this
October 23, 2026
Compromised LiteLLM 1.107.0 on PyPI Installs a Credential Stealer and Backdoor safedep.io safedep.io
2 people saved or upvoted this
October 22, 2026
DeepInfer: High-performance inference framework for DeepSeek models github.com github.com
1 person saved or upvoted this
If you don't run your own LLM proxy, how do you know what's going on? jrdsgl.com jrdsgl.com
2 people saved or upvoted this
Agentic Monitoring benjamincongdon.me benjamincongdon.me
2 people saved or upvoted this
October 21, 2026
Hyper Optimizing Kimi K3 in vLLM vllm.ai vllm.ai
1 person saved or upvoted this
October 20, 2026
How the Claude Code team does code review claude.dev claude.dev
1 person saved or upvoted this
October 16, 2026
Show HN: Parsimony – an open-source benchmark for how much code AI models add parsimony.lasu.dev parsimony.lasu.dev
1 person saved or upvoted this
Pensieve (formerly tedhaley.ca/pensieve): local indexing to reduce tool calls and token usage pensieve.tools pensieve.tools
1 person saved or upvoted this
October 15, 2026
Codex CLI installs a persistent, self-updating background daemon by default github.com github.com
1 person saved or upvoted this
I posted to HN, and then my website was flooded with AI bot traffic sammystraus.com sammystraus.com
1 person saved or upvoted this
October 14, 2026
How to serve trillions of tokens for trillion-parameter coding agents modal.com modal.com
1 person saved or upvoted this
Janas: Run LLMs larger than RAM by streaming experts from SSD github.com github.com
2 people saved or upvoted this
October 13, 2026
500B tokens later: Letting AI agents decompile a first-person shooter momo5502.com momo5502.com
2 people saved or upvoted this
October 12, 2026
Hitting 1B tokens/minute on 1 GPU combining a query planner and inference engine modal.com modal.com
1 person saved or upvoted this
Local LLM Inference at Scale with vLLM data4sci.substack.com data4sci.substack.com
1 person saved or upvoted this
October 11, 2026
Project Arena: open Kubernetes incident benchmark for AI SRE investigation tools github.com github.com
1 person saved or upvoted this
Tokens to GPUs: a planning calculator with its formula and sensitivity analysis exposed cedana.com cedana.com
1 person saved or upvoted this
October 10, 2026
LittleBit: Sub-1-Bit LLM Compression via Latent Factorization github.com github.com
1 person saved or upvoted this
Show HN: Rgpu – a PyTorch device whose tensors live on a remote GPU github.com github.com
1 person saved or upvoted this
October 09, 2026
OpenAI "rogue" agent activities found on Wikimedia projects diff.wikimedia.org diff.wikimedia.org
2 people saved or upvoted this