What “no credit card” means here
This comparison includes 23 of the 25 providers with detailed pages. Each one documents a path to API access before a payment method is verified. That condition says nothing by itself about how long the allowance lasts: a permanent provider tier, a zero-priced model pool, and one-time signup credit all qualify for this page but remain separate access types in the table. A service that merely delays billing after requiring a card does not qualify.
No-card providers and their published limits
The 23 rows below are not ranked and no position is sponsored. Published numbers appear only when the provider states them in an official source; dynamic, per-model, or console-only allowances stay descriptive instead of being converted into guessed requests per minute. Open the provider page before signing up to see its exact model list, lifecycle note, source URLs, and last review date.
| Provider | Free access | Published limits | Card | Base URL |
|---|---|---|---|---|
| Google Gemini API | Provider free tier | Set by project tier | Not required | https://generativelanguage.googleapis.com/v1beta/openai/ |
| GroqCloud | Provider free tier | 30 requests per minute, 1000 requests per day | Not required | https://api.groq.com/openai/v1 |
| SambaNova Cloud | Provider free tier | 20 requests per minute, 20 requests per day | Not required | https://api.sambanova.ai/v1 |
| Cohere | Provider free tier | 20 requests per minute | Not required | https://api.cohere.ai/compatibility/v1 |
| Cloudflare Workers AI | Provider free tier | Published in compute units | Not required | https://api.cloudflare.com/client/v4/accounts/ACCOUNT_ID/ai/v1 |
| Hugging Face Inference Providers | Provider free tier | Published as a credit balance | Not required | https://router.huggingface.co/v1 |
| SiliconFlow | Provider free tier | 1000 requests per minute | Not required | https://api.siliconflow.com/v1 |
| Fireworks AI | Provider free tier | 10 requests per minute | Not required | https://api.fireworks.ai/inference/v1 |
| Z.AI Open Platform | Provider free tier | Selected models priced at zero | Not required | https://api.z.ai/api/paas/v4 |
| Novita AI | Provider free tier | Selected models priced at zero | Not required | https://api.novita.ai/openai |
| Mistral La Plateforme | Provider free tier | Enforced but not published | Not required | https://api.mistral.ai/v1 |
| Alibaba Cloud Model Studio | Provider free tier | Published per model | Not required | https://dashscope-intl.aliyuncs.com/compatible-mode/v1 |
| Moonshot AI (Kimi) | Provider free tier | Published per tier | Not required | https://api.moonshot.ai/v1 |
| Pollinations.AI | Provider free tier | Enforced but not published | Not required | https://text.pollinations.ai/openai |
| Ollama Cloud | Provider free tier | Enforced but not published | Not required | https://ollama.com/v1 |
| Vercel AI Gateway | Free trial credit | Published as a credit balance | Not required | https://ai-gateway.vercel.sh/v1 |
| IBM watsonx.ai | Free trial credit | Enforced but not published | Not required | https://us-south.ml.cloud.ibm.com/ml/v1 |
| OpenRouter | Free model aggregator | 20 requests per minute, 50 requests per day | Not required | https://openrouter.ai/api/v1 |
| Together AI | Metered access | Dynamic, no fixed numbers | Not required | https://api.together.xyz/v1 |
| Nebius Token Factory | Metered access | 60 requests per minute | Not required | https://api.tokenfactory.nebius.com/v1 |
| Perplexity API | Metered access | 50 requests per minute | Not required | https://api.perplexity.ai |
| DeepInfra | Metered access | Enforced but not published | Not required | https://api.deepinfra.com/v1/openai |
| Chutes | Metered access | Published per paid plan | Not required | https://llm.chutes.ai/v1 |
Choose by repeatability, not just signup friction
If you need a hobby project to keep running, start with a provider free tier and confirm that the specific model is included. If you only need to evaluate an integration, trial credit can be enough, but plan what happens after the balance expires. Aggregators can offer broader model choice while assigning limits per model. A no-card label removes one signup barrier; it does not promise unlimited use, anonymous registration, production support, or a tier that will never change.
The trade-offs hidden behind the easy signup
Providers commonly control free usage through request caps, token caps, model queues, regional availability, or account-level abuse checks. Two services can both say “free” while one publishes a stable daily request count and the other adjusts capacity dynamically. Compare the limit wording and model ids together. A high RPM number is less useful when the model you need is excluded, and a large daily allowance may still feel slow when concurrency is restricted.
A safe way to test one
Create the key on the provider’s own signup page, store it in an environment variable, and begin with the smallest documented model. Send one request before changing a client configuration so credential errors and protocol errors remain easy to distinguish. For OpenAI-compatible services, change the Base URL and model id while keeping the standard SDK. Never paste a credential into an issue, commit it to a config file, or treat a public key as shared quota.
How this shortlist stays verifiable
Payment requirements and limits were last checked against provider-owned sources on 2026-07-25. The methodology explains why missing numbers stay missing, and the browser key checker can classify your own request as accepted, rejected, limited, or unreachable without sending the key to this project. A sample response is never presented as proof of provider-wide uptime.