Free LLM Radar live · OpenRouter free tier
loading…
—
Model Providers Best uptime 24h Context Capabilities Tokenizer Knowledge
awaiting first refresh — the OpenRouter catalog has not been fetched yet on this server.
—
Provider Free models served Avg uptime 24h Total context (sum tokens) Tier

Free LLM Radar

A live dashboard of every model on OpenRouter that costs nothing to call. We fetch the public OpenRouter catalog every 15 minutes and the per-model endpoint stats every hour, persist the full snapshot in SQLite, diff each new snapshot against the previous one, and surface the deltas as a drift feed. All data is real, all sources are public, no key required.

Sources

  • GET https://openrouter.ai/api/v1/models — every 15 min
  • GET https://openrouter.ai/api/v1/models/{id}/endpoints — every hour

What counts as "free"

A model is included if its public OpenRouter pricing has both prompt and completion set to exactly "0". Models in trial, in preview, or rate-limited are still listed as free if they meet that criterion.

What we don't do

  • No mocked, seeded, or fallback model lists. If the API is down, the UI is honest about it.
  • No keys, no auth, no accounts.
  • We do not call inference. We describe; we never invoke.

Reliability score

For each model, the reliability score is the maximum 24-hour uptime across its providers. Rationale: a developer only needs one working route, so the model is effectively as reliable as its best provider.

free-llm-radar · v1 · built by Cowork (Claude Opus 4.7)