Skip to content

7.4 billion tokens per month. 34 free LLM providers. 635 free model endpoints. All behind one /v1 endpoint, plus any custom OpenAI-compatible endpoint. Smart routing, automatic failover, encrypted keys. Personal experimentation only.

TypeScript
3.2 KB
Raw Blame

LLMTR

LLMTR exposes an OpenAI-compatible API at https://llmtr.com/v1. Create an account, verify email, and create a dashboard API key. Add it under Keys → LLMTR, or import LLMTR_API_KEY from an environment file. Never commit the key.

Selected routes are zero-priced with daily or rolling quotas, not monthly cash credits. The migration guide documents free requests at zero balance. Account caps can vary; this integration does not invent numeric quotas or make the whole provider roster free. New accounts have a reduced free allowance that increases automatically once the account is 24 hours old. A quota response observed during testing confirmed this waiting option; no top-up is required. A streaming smoke test reached that quota, so only non-streaming inference was live-verified in this publication.

Catalog and release timing

Model rows are supplied only by the hosted signed catalog. There are no bundled seeds or automatic discovery of LLMTR's paid models. Existing Premium-immediate and Free-after-30-days release logic is unchanged. Older app versions need an update to recognize the new provider.

Five routes returned matching model IDs and valid text completions on 2026-09-29:

  • qwen/qwen3.8-27b-free
  • nvidia/nemotron-3-ultra-550b-a55b
  • nvidia/nemotron-3-super-120b-a12b
  • motif/motif-3
  • inclusionai/ling-3.0-flash-fin

These checks are not independent proof of the underlying model weights. Tools, vision and context limits require separate verification before catalog flags are enabled. Motif's quota replenishes each request after 24 hours. Expiring promotions and EVREN BYOK routes are excluded.

Authentication and errors

The public /v1/models endpoint cannot validate a key. Dashboard /api/usage rejected API-key authentication during testing. Instead, key validation submits a deliberately nonexistent model without generating tokens. Live valid/invalid controls returned 404/model_not_found and 401/auth_error, respectively. Other errors remain inconclusive; quota exhaustion does not mark a key invalid.

Key-validation pacing

LLMTR publishes no numeric rate limit for this endpoint, but a burst of the same nonexistent-model probe from one IP is the pattern their security system reads as automation, and an account flagged that way gets suspended — the report on #1369 came from an account that was already locked out and learned of it from the provider's own email. Probes are therefore paced: they run one at a time, and concurrent checks of the same key share a single request. The gap between the start of two probes defaults to 2 seconds and is tunable with LLMTR_VALIDATION_MIN_GAP_MS (0 disables it). This covers every caller of the probe — the 5-minute health pass, the dashboard's "check all" button, the post-wake re-probe, the cooldown-probe job and the 401-triggered revalidation. Chat and streaming traffic is never paced.

Chat and streaming use the shared OpenAI-compatible transport, preserve usage and backoff, and reject mismatched response model IDs rather than silently accepting substitutions. No automatic paid fallback is introduced by this adapter.