LLM Inference & Serving

The Unified Interface For LLMs

Easy, fast, and cost-efficient LLM serving for everyone.

High-performance serving framework for large language and multimodal models.

Go from prompt to production with Gemini, Veo, Nano Banana, and more

The AI community building the future.

Build with open models, on your computer and in the cloud

LLM inference in C/C++

Build leading AI products on OpenAI’s platform

Bring Grok into your terminal.

The AI Native Cloud

Easily run & train models locally.

The platform for building generative AI applications and agents at production scale

Generative media platform for developers.

The fastest inference on any silicon

Inference Platform: Deploy AI models in production

Meet your thinking partner

Block Diffusion for Flash Speculative Decoding

The AI Gateway for developers

Run local AI models on your device

Low-latency, real-time voice and vision interactions with Gemini.

Everything to power your Hermes Agent

Designed for rapid, reliable deployment of accelerated generative AI inference anywhere.

A Unified, Open Source RL Backend for Enterprise Post-Training

An enterprise-grade, all-in-one large model platform centered on Agent development

Develop with Claude

Access Claude's full platform capabilities through AWS with Anthropic-managed infrastructure.

State-of-the-art diffusion models for image, video, and audio generation in PyTorch.

Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation

Automatically select the best model for each task.

Fast inference and fine-tuning for open source models

High-performance linear attention kernel library built on TileLang

Discover, download, and run local LLMs

Build with Muse Spark on Meta Model API - now in public preview for US developers.

Enterprise-Grade Reinforcement Learning for Large-Scale Model Training

Inference Built for Coding Agents

A new beginning.

Guarantee long-term access to OpenAI compute for the products, agents, and customer workflows that matter most.

Latency-first serving engine for Physical AI

The Open Superintelligence Stack

Save 50% on cost with the same capability and behavior.

Optimized GPU kernels for LLM operations, built with TileLang.

Easy, fast, and cheap omni-modality model serving for everyone

A full-stack platform for open superintelligence.

Inference built for agentic intelligence.

Continual learning and adaptation after deployment.

AFM model playground

Production AI agents

One API for 1000+ AI models

Building Open Intelligence

A SuperPoD AI infrastructure option for large-scale AI training and inference.

LLM API for Baidu AI Studio

AI-Native Cloud for Enterprise Growth

Get instant access to inference that’s up to 15x faster than NVIDIA GPUs.

Manage every agent and provider from one place.

Best Subscription for Open Weight Models.

Run AI models on Cloudflare's global network.

Comprehensive LLM serving solution for Huawei CloudMatrix384

The best coding agents run on Baseten

Software product by Conway Research

Bring PyTorch models to Core AI for on-device execution.

Platform for Accelerated Computing

Data and AI for all

Ship AI that keeps getting better.

Harness the power of DeepSeek R1's reasoning and Claude's creativity and code generation capabilities with a unified API and chat interface.

DeepEP: an efficient expert-parallel communication library

clean and efficient BLAS kernel library on GPU

Model architecture summarizer for AI/ML model files

DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm

Run frontier AI locally.

FIne tuning with NO labels

Bringing Flash Magic to Classical Machine Learning Operators

Efficient Multi-head Latent Attention Kernels

Lightning-Fast Model Loading for PyTorch

An LLM Pretraining Framework Built End-to-End by an Autonomous Agent Loop

Pure-Rust, CPU-only OCR engine with a five-model zoo and no Python, CUDA, ML framework, or GPU required at inference.

Innovate, build, and deploy enterprise ready agents

Speak it. See it. Share it.

AI Coding Powered by GLM-5.2 & GLM-5-Turbo for Agents & IDEs

Explore, Experience, and Evaluate the Future of On-Device Generative AI with Google AI Edge.

Private & Local AI Chatbot

Fast LLM inference, OpenAI-compatible. Simple to integrate, easy to scale.

Use Garry Tan's exact Claude Code setup: 23 opinionated tools that serve as CEO, Designer, Eng Manager, Release Manager, Doc Engineer, and QA

SLO-aware Benchmarking and Evaluation Platform for Optimizing Real-World LLM Inference

The platform for scalable, reliable, and efficient inference

A model built for deterministic tasks

The Universal Gateway for AI Inference

Control every model call

Idea to AI product, ⚡️ fast.

Open-Source AI Gateway & LLM Proxy

LM Studio’s headless daemon

Use your local models, remotely.

Supercharge Your LLM with the Fastest KV Cache Layer

Augment your apps with AI

Your AI production platform.

One API for All AI Media Generation

Model-as-a-Service

Inference reimagined, from Kernel to Cloud. One unified stack.

A KVCache-centric Disaggregated Architecture for LLM Serving.

Multimodal AI software product

Ship Apps. Agents. Companies. One prompt. One config file. All your infrastructure.

Next-Generation LLM Gateway and AI Asset Management System

Runtime for AI agents to execute real-world tasks

The foundation for your AI factory.

Designed to build and run autonomous agents.

A Scalable and Efficient Post-Training Library

4-bit floating-point format and low-precision recipe for efficient AI training and inference.

Free AI Image & Video Studio

AllenAI's post-training codebase

Open-source specification and ecosystem for multi-provider, interoperable LLM interfaces.

Open-source theoretical reconstruction of the Claude Mythos Recurrent-Depth Transformer architecture

2-10x accuracy improvements on reasoning tasks with zero training

One AI gateway: adaptive LLM routing & governance

AI Toolkit

One-stop AI teaching and training platform

Fast speech recognition with NVIDIA's Parakeet models in pure C++.

Model Routing, Adaptive Inference & Fine-Tuning

Machine Learning Platform for AI

A TTS that fits in your CPU (and pocket)

Prime Intellect's open research platform for post-training

PrismML platform

Define functions in English. Run them locally.

Tile primitives for speedy kernels

High-performance, scalable framework for deploying Large Language Models (LLMs) in production.

DeepSeek-native coding agent for your terminal

Advancing biological preparedness with trusted developers and government partners.

The AI Developer Cloud

Edge AI platform

Open Medical Superintelligence

Launch, manage, and stop LLM inference workloads on one or more NVIDIA DGX Spark systems — no Slurm, no Kubernetes, no fuss.

From Coding to Agents, Build It All

Amortized KV Cache Compaction in a Single Forward Pass

The first model built for long-context tasks

The first model built for long-context tasks

Carrier-owned global communications platform for voice AI agents, SIP trunking, programmable voice, SMS/MMS, and AI inference.

AI Inference at the speed you want and the margin you need

Compute for every scale.

Thinking Machines software platform

A concise domain-specific language designed to streamline the development of high-performance GPU/CPU kernels.

Ascend TileLang adapter

Fine-tune open-source models for real production use

Train Dreambooth models. Generate unlimited AI assets.

Easily run & train models locally.

GPU cloud marketplace for renting and hosting GPU compute

The Only Private AI API

Production inference for encoders, poolers, and structured prediction — as vLLM plugins.

Workers AI provider for the AI SDK.

Unified API for 100+ AI Models

一站式大模型开发与应用平台