AI & Tech
Freebies

NVIDIA API Catalog

Hosted NVIDIA NIM endpoints for trying open models without owning a GPU.

Documentation checked 20 Oct 2026. Not personally tested.

What you get

Free hosted inference, rate-limited.

Renewal
Not stated; rate limits apply per account
Credit card
Card not confirmed
Who can use it
NVIDIA developer account; free access is intended for prototyping and evaluation.
Expiration
No end date stated; access is subject to rate limits and model availability.
Access type
LLM APIs, Image generation

The catch

Free access is throttled and shared: rate limits vary by model, and other users can slow you down at peak times. Production workloads are expected to move to a dedicated NIM deployment or a licensed NVIDIA AI Enterprise setup.

Before you claim it

  1. Open the provider’s official page and confirm your account and region qualify.
  2. Check the live allowance for the exact model or feature you intend to use.
  3. Review billing settings before enabling paid usage or adding a payment method.

Official sources

Referral transparency

Affiliate availability has not been established. This listing uses a direct provider link.

Read our disclosure

Information changed?

Provider terms can change after a check. Follow the official sources for the latest account-specific terms.

Keep exploring

Groq

LLM APIs · Voice & audio

GPT-OSS 120B: 200k tokens/day

Fast hosted inference for language and speech models, with a recurring Free plan.

No credit cardRecurring allowance

The catch. GPT-OSS 120B also has limits of 30 requests/minute and 8,000 tokens/minute. Limits differ by model and apply at organization level. Upgrading to Developer enables paid usage.

Docs checked 20 Oct 2026

OpenRouter

LLM APIs

50 requests/day across free models

One API for trying multiple free language models.

No credit cardRecurring allowance

The catch. 20 requests/minute. Failed attempts count toward the daily limit. Paid models require credits; free availability is not a production SLA. Check the selected model's data policies.

Docs checked 20 Oct 2026