Providers

Cerebras

Provider

Wafer-scale inference engine. Serves frontier open-weight models at 2000+ tok/sec.

4 active deals

Cloudflare Workers AI

Cloud

Edge-deployed inference from inside Cloudflare Workers. Free allowance of compute units (Neurons) every day.

1 active deals

Cohere

Provider

Enterprise-focused LLM provider with a non-commercial trial tier covering chat, embeddings, and reranking.

1 active deals

GitHub Models

Cloud

Free experimentation tier for OpenAI, Meta, Mistral, and xAI models, authenticated with a GitHub personal access token.

1 active deals

Google AI Studio

Cloud

Google's AI playground with a generous free tier for Gemini models — with a data-training catch.

1 active deals

Groq

Provider

Ultra-fast inference on custom LPU hardware. Runs open-weight models at hundreds of tokens per second.

3 active deals

Hugging Face

Aggregator

Open model hub with a routed Inference Providers API covering 200+ models. Free tier is symbolic.

1 active deals

Microsoft Azure

Cloud

Microsoft's cloud platform. Hosts Azure OpenAI Service with GPT-4o, GPT-4.1, GPT-5 and the o-series, plus Azure AI Foundry models.

1 active deals

Nous Research

Provider

Research lab behind the Hermes and DeepHermes model families. Runs Nous Portal — an OpenAI-compatible API with signup credits.

1 active deals

NVIDIA NIM

Cloud

100+ models behind one OpenAI-compatible endpoint. Free inference credits for NVIDIA Developer Program members.

1 active deals

OpenRouter

Aggregator

Aggregator that routes to 200+ models across providers through a single OpenAI-compatible API. Hosts free variants of many models.

1 active deals

SambaNova Cloud

Provider

Inference on custom RDU silicon. One of the few providers serving 405B Llama models free on signup credits.

1 active deals