LLM APIs from OpenAI, Anthropic, Google, Together AI and more
24 tools in this category
The original LLM API, home to the GPT-5 series.
API for Claude Sonnet and Opus models with long context.
API for Gemini 2.5 and 3.x models, up to 1M token context.
Inference platform for 100+ open-source LLMs at scale; free models plus paid API.
Serverless inference for fast, low-cost LLM serving.
The standalone per-token Endpoints LLM API was folded into the Anyscale platform, which now sells GPU compute for Ray workloads; Anyscale signed a definitive agreement to join Nscale in July 2026 - see the current alternatives guide.
Ultra-fast inference on custom LPU hardware; free tier plus paid API.
Enterprise-grade LLM for retrieval-augmented generation.
Cut about 60% of its staff in May 2026 and stopped selling standalone Jamba models; the Jamba and Maestro APIs were sunset on 9 August 2026 as AI21 moved into custom model training - see the current alternatives guide.
Serverless inference for 100k+ open models.
Run open-source AI models with a simple API.
Serverless infrastructure for AI inference at scale.
