AI model APIs

27 services

Free inference tiers are limited by requests per minute and per day, and the numbers change often. Google AI Studio allows 20 requests a day and 5 a minute on Flash models, and 14,400 requests a day on Gemma. OpenRouter offers some open models at no cost with rate limits, onomeo allows 60 requests per 5 hours across 35 free models, and Gonka Broker includes 1M tokens a month. Pollinations.AI generates images without an account. Prompts sent on free tiers may be used to improve the provider's models, so read the data terms before sending anything sensitive.

The category also covers the tooling around models. Gateways such as Portkey and Keywords AI each route 10,000 requests a month with logging. For tracing and evaluation, Langfuse includes 50,000 observations a month and is open source, Arize AX 25,000 spans with 15-day retention, telemetry.dev 10,000 spans with 7-day retention, and Braintrust 1,000 eval rows a week. Search APIs built for agents include Tavily AI with 1,000 requests a month and Brave Search API with a $5 monthly credit.

27 services

App AI Gateway

Proxy that lets iOS apps call AI providers without exposing keys, verified with App Attest.

  • No card

1,000 requests/mo

Visit
Arize AX

Tracing and evaluation platform for LLM apps and agents, with self-hosted Phoenix option.

  • No card

25,000 spans/mo

1 GB ingestion/mo, 15-day retention, Unlimited users

Visit
Braintrust

LLM evaluation suite with a prompt playground and dataset management.

1,000 eval rows/wk

Visit

Web, news, image and video search API from an independent index, usable for RAG.

$5 credit/mo

Visit
Future AGI

Open-source evals, tracing, agent simulation and guardrails for LLM apps.

  • Open source

100,000 gateway requests/mo

50 GB storage, 2,000 eval credits, Unlimited seats

Visit
Gonka Broker

OpenAI-compatible inference for open models running on a decentralized GPU network.

1M tokens/mo

Visit
Google AI Studio

Browser workspace and API keys for prototyping with Gemini and Gemma models.

20 requests/day (Flash)

5 requests/min (Flash), 14,400 requests/day (Gemma)

Visit
Hugging Face

Hub for sharing ML models and datasets, with hosted inference and training tools.

30,000 input chars/mo

Visit
Keywords AI

LLM gateway with monitoring that routes calls to hundreds of models through one API.

10,000 requests/mo

Visit
Langfuse

Open-source tracing, prompt management and evals for debugging LLM applications.

  • Open source

50,000 observations/mo

Visit
LangWatch

LLMOps monitoring and evaluation with DSPy-based prompt optimization.

  • Open source

1,000 traces/mo

1 DSPy workflow

Visit
Latitude

MIT-licensed tracing and evaluation for AI agents running in production.

  • Open source

20,000 credits/mo

30-day retention, Unlimited seats

Visit
Lumenfall.ai

OpenAI-compatible gateway to image generation models with provider failover.

Unlimited FLUX.1 schnell images

Visit
Maxim AI

Evaluation, simulation and observability for LLM agents, including a prompt playground with BYOK.

10,000 logs/mo

3 seats

Visit
Mediaworkbench.ai

Workspace for chatting with Azure OpenAI, DeepSeek and Gemini models, including images.

100,000 words

Visit
onomeo

Single OpenAI-format key for many models, usable from Cline, SillyTavern or Open WebUI.

35 free models

60 requests per 5 hrs

Visit
Portkey

AI gateway with request logging and observability for generative AI apps.

10,000 requests/mo

Visit
Tavily AI

Web search API built for research tasks, returning organized results for apps and agents.

  • No card

1,000 requests/mo

Visit
telemetry.dev

OpenTelemetry tracing for LLM apps recording tokens, cost, latency and tool calls.

  • No card

10,000 spans/mo

7-day retention, 1 project, 2 seats

Visit
Transcript LOL

Audio and video transcription with LLM summaries, from uploads or social media links.

2 transcriptions/day

Visit
Comet Opik

Open-source tool for tracing, testing and evaluating LLM applications.

  • Open source

Limits not listed

Visit
Composio

Tool integrations and auth layer that connects AI agents to 200+ apps.

Limits not listed

Visit
OpenRouter

Unified API for many LLMs, some open models offered at no cost with rate limits.

Limits not listed

Visit
Pollinations.AI

Open-source image generation API callable without an account or key.

  • No sign-up
  • Open source

Limits not listed

Visit
ReportGPT

AI writing assistant for drafting reports using your own model API key.

Limits not listed

Visit
Wolfram Language

Cloud access to the Wolfram Language and its curated computational knowledge.

Limits not listed

Visit
Esc

Type to search every entry. ↑↓ to move, Enter to open.