dev-tools
llmspeed: Feel Local LLM Speed Before You Buy Hardware
llmspeed simulates tokens per second and time-to-first-token for local LLMs across 30 GPUs, Macs and CPUs, so you can judge hardware before buying.
#dev-tools#local-llm#benchmark
Tag
GPU Posts
19 posts tagged #gpu
Browse 19 posts tagged GPU. Setup notes, reviews, comparisons, and workflow patterns.
All Posts
ai-setup
PyTorch: Tensors and Dynamic Neural Networks in Python
PyTorch is the open-source deep learning framework built around tensors and a dynamic computation graph, with first-class GPU acceleration. Learn what it does and when to use it.
#pytorch#machine-learning#deep-learning
dev-tools
DeepGEMM: a clean GPU BLAS kernel library
DeepGEMM is a Cuda BLAS kernel library from DeepSeek for efficient GPU matrix math. MIT licensed, 8.7k stars, 1.3k forks.
#github-trending#cuda#gpu
dev-tools
tilelang: a DSL for GPU kernel development
Tile Language is a Python-embedded DSL aimed at writing high-performance GPU, CPU, and accelerator kernels with less boilerplate. 8,121 stars.
#github-trending#gpu#kernels
dev-tools
Lambda CLI – MCP Server and GPU Instance Manager
Unofficial CLI and MCP server to manage Lambda cloud GPU instances directly from your terminal or any MCP-compatible AI assistant.
#mcp#cli#gpu
dev-tools
PrettyMux - GPU-accelerated Linux terminal multiplexer
PrettyMux is a native Linux terminal multiplexer built with GTK4 and libadwaita — GPU-accelerated with AI agent-friendly notifications, split panes, workspaces, and prettymux-open CLI.
#linux#terminal#dev-tools
self-hosted
dstack - Unified GPU Orchestration Across Clouds and
dstack is an open-source control plane for provisioning and orchestrating GPU workloads across NVIDIA, AMD, TPU, and Tenstorrent on any cloud, Kubernetes, or bare metal.
#gpu#orchestration#ml-ops
dev-tools
Lambda CLI – MCP Server for Lambda GPU Cloud
A fast CLI and MCP server for managing Lambda cloud GPU instances via natural language commands in Claude Code or other AI assistants.
#mcp#gpu#cloud
ai-setup
Velda – Run GPU Batch Jobs Without Docker or Manifests
Velda turns any dev environment command into a cloud batch job. No Dockerfiles, no image registries, no Kubernetes manifests — just prefix with vrun and scale.
#ai-setup#dev-tools#gpu
ai-setup
Shadeform GPU Cloud Marketplace - Compare 30+ Providers
Access H100, H200, B200 GPUs on demand from 30+ cloud providers via a single API. Compare prices, deploy clusters up to 64 GPUs, no long-term commitment.
#gpu#cloud-computing#ai-infrastructure
dev-tools
Chamber (YC W26) – AI Agent for GPU Infrastructure
Chamber is an AIOps platform that puts an AI agent between your team and your GPU fleet. Ask in plain language, get root-cause analysis and automated remediation across Kubernetes and Slurm.
#gpu#ai-agent#dev-tools
dev-tools
Lambda CLI - MCP Server for Lambda Cloud GPU Instances
Unofficial CLI and MCP server for managing Lambda cloud GPU instances from the terminal or inside AI editors like Cursor and VS Code.
#dev-tools#mcp#gpu
ai-setup
Chamber – AIOps AI Teammate for GPU Infrastructure
Chamber deploys an AI agent named Chambie that monitors, root-causes, and fixes GPU issues across AWS, GCP, Azure, and on-prem clusters — answering in Slack. YC W26.
#aiops#gpu#infrastructure
dev-tools
Koyeb - Serverless Deploy for APIs and AI Inference
Deploy APIs, AI inference, and databases on Koyeb's global serverless infrastructure. GPU instances, per-second billing, autoscaling to zero. CLI setup guide.
#serverless#deployment#gpu
ai-setup
IonRouter - High-Throughput Inference API
YC W26 inference API with custom IonAttention engine. Drop-in OpenAI replacement, built for NVIDIA Grace Hopper. Per-second billing, no cold starts.
#ai-inference#yc-w26#gpu
dev-tools
OpenLake: GPU-Native Storage in Rust
High-throughput GPU storage engine built on Rust and io_uring — GPUDirect Storage, RDMA, erasure coding, 6x higher throughput than conventional object stores.
#github-trending#dev-tools#rust
dev-tools
Expanse – HPC GPU Cluster Resource Optimizer
Expanse reads your job's source code, submission script, and hardware telemetry to predict actual GPU needs—cutting 59% datacenter waste before jobs run.
#hpc#gpu#ai-infrastructure
dev-tools
Chamber - AI Teammate for GPU Infrastructure
AI agents that autonomously monitor, root-cause, and remediate GPU infrastructure issues. Reduce compute costs, improve GPU utilization, and accelerate ML research. YC W26.
#ai-agents#gpu#infrastructure
ai-setup
Chamber – AI Teammate for GPU Infrastructure
Chamber is a YC W26 AI teammate that autonomously monitors, root-causes, and remediates GPU infrastructure issues across multi-cloud.
#ai-ops#gpu#infrastructure