Free LLM API Status & Real-Time Uptime Monitor

Track real-time uptime, availability, and success rates for popular free AI providers like Groq, OpenRouter, and Gemini.
Powered by real user tests — not synthetic pings.

Live API Health & Availability

Mostly Operational
9/13 providers healthy
Last 24 Hours • Excludes user configuration errors

Google Gemini API Status

11.1% uptime

Last 24 hours: 11.1% available across 9 community tests.

LLM7.io API Status

100% uptime

Last 24 hours: 100% available across 7 community tests.

NVIDIA NIM API Status

100% uptime

Last 24 hours: 100% available across 6 community tests.

GitHub Models API Status

0% uptime

Last 24 hours: 0% available across 6 community tests.

Groq API Status

100% uptime

Last 24 hours: 100% available across 5 community tests.

Ollama Cloud API Status

100% uptime

Last 24 hours: 100% available across 5 community tests.

OpenRouter API Status

100% uptime

Last 24 hours: 100% available across 5 community tests.

Z AI (Zhipu AI) API Status

100% uptime

Last 24 hours: 100% available across 5 community tests.

ModelScope API Status

80% uptime

Last 24 hours: 80% available across 5 community tests.

Cohere API Status

0% uptime

Last 24 hours: 0% available across 5 community tests.

Agnes AI API Status

100% uptime

Last 24 hours: 100% available across 1 community test.

Cloudflare Workers AI API Status

100% uptime

Last 24 hours: 100% available across 1 community test.

Mistral AI API Status

100% uptime

Last 24 hours: 100% available across 1 community test.

Frequently Asked Questions (FAQ)

Is the OpenRouter or Groq API down? How do you track uptime?

Unlike traditional status pages that rely on predictable synthetic pings from identical IP addresses, this dashboard is powered entirely by real-world API requests from our community users. When users test their keys in the Dashboard, the results are anonymously aggregated here. This means our uptime percentages reflect actual user experience, including rate limits and geographical routing issues that typical pings might miss.

What do the API rate limits (429) or balance errors (402) mean for uptime?

We do not count user-side configuration errors against a provider's health score. Failures caused by invalid API keys (HTTP 401/403), insufficient account balance (HTTP 402), or hitting personal rate limits (HTTP 429) are automatically filtered out. Only true internal server errors (HTTP 500+), gateway timeouts, and total unavailability will cause the API status to be marked as degraded or down.