AI & machine learning

LLM routing: chat, embeddings, images, audio and batch inference across models, plus key and credit management

AI chat answers: ask ChatGPT questions and get Markdown replies with cited web links

Model serving: manage models, subscriptions, API keys, access policies and external providers

AI model hosting: upload model versions, deploy them to endpoints, and check their status

LLM gateway: chat, completions, embeddings, images, audio, fine-tuning, files, assistants and vector stores

AI characters: generate chat completions and fetch a character's greeting message

AI coding agents: launch, follow up on, stop and download work from repositories

AI models: manage local model files, aliases, remote API models, MCP servers, tokens and settings

AI agents: run threads, issues, Captains, automations, files and review rules

Hosted AI agents and sandboxes: deploy, run and wire agents to messaging, email and billing

AI generation: create images, video, 3D models and audio, plus enhance and upscale media

LLM tracing and evaluation: log calls, traces, datasets, scorers, feedback and model completions

AI agents: manage agent workspaces, tools, scheduled tasks and conversation invocations

AI models: generate text and speech, transcribe audio, embed text, and manage files and batches

AI models: generate text, speech and embeddings, manage files, batches and fine-tuning jobs

AI gateway: route OpenAI-compatible chat and audio calls, manage keys, budgets and usage

Agent infrastructure: manage sessions, sandboxes, disks, artifacts, agent skills, users and projects

Video and media AI: upload files, extract data, transcribe, describe, search and chat over collections

Audio: generate sound effects for video and text, extend and inpaint clips, transcribe audio to MIDI

AI agent workspace: manage actors, files, skills, meetings, approvals, memory and integrations

Video and image generation: create jobs, price them, upload references, track balance and usage

LLM gateway: chat, embeddings, images, video, audio, key management and usage metrics

Datasets, training, models and deployments: train YOLO on cloud GPUs, deploy endpoints, export to ONNX or TensorRT

Aggregated APIs for flights, prediction markets, crypto prices and weather, callable by AI agents