AI dev and agent tools, operator-tested.

Direct API access to Claude Opus, Sonnet, and Haiku model families.

API access to Mistral and Codestral models, plus la Plateforme tooling.

API access to GPT models, embeddings, audio, and fine-tuning endpoints.

Fast, low-cost inference for open and proprietary models with native function calling.

Sub-100ms LPU-based inference for Llama, Mixtral, and other open models.

Inference and fine-tuning across 200+ open-source LLMs with serverless and dedicated endpoints.

Unified OpenAI-compatible API for 200+ LLMs with provider routing and failover.

AI gateway with prompt versioning, caching, governance, and 50+ provider integrations.

OpenAI-compatible gateway with provider routing, observability, and cost tracking.

Production-grade managed vector database with serverless and pod-based tiers.

Open-source vector database with rich filtering, payload indexing, and quantization.

Role-based multi-agent orchestration framework with task delegation and tool use.

Open-source framework for LLM applications, agents, and RAG pipelines.

TypeScript framework for agents, workflows, and RAG with first-class evals.

Open-source AI code-assistant for VS Code and JetBrains; bring-your-own-model.

Open-source AI pair programmer; edits via diffs and commits as it works.

Anthropic's official CLI for autonomous coding — runs in your terminal, edits your repo.

Autonomous coding agent inside VS Code; reads, plans, edits, and runs in a sandboxed shell.

AI-native IDE built on a fork of VS Code; agent mode, tab completion, and inline edits.

Open-source LLM observability with built-in caching, gateway, and rate limiting.

Open-source LLM observability, evals, prompt management, and dataset tooling.

Local LLM runtime for macOS, Linux, and Windows with a built-in model library.