Evolink

EvoLink is an AI model API platform for developers building applications, agents, or media-generation workflows. The official site presents a single access layer for language, image, video, and audio models, with model pages and documentation to help select an integration.

Integrating model access

EvoLink describes compatibility with OpenAI, Anthropic, and Google request formats. Its setup sequence involves creating an account, generating an EvoLink API key, choosing an integration format, and updating the base URL and key in the application. A developer should still inspect the selected model's documented inputs and outputs before relying on compatibility across a whole workflow.

The site describes automatic routing and provider failover, alongside dashboard usage and cost monitoring. Its uptime and routing-latency figures are vendor claims rather than measurements established here. The listed model counts also differ between the product and pricing pages, so consult the current model catalogue for a specific model instead of choosing the service by a headline count.

Prepaid usage and model pricing

EvoLink uses prepaid, pay-as-you-go billing without a subscription or seat fee. The pricing page lists a minimum top-up of $10 and says purchased credits do not expire. Promotional credits may have their own expiry date.

Charges depend on the model and billing unit. Language models use token pricing, image models use per-image pricing, and video or audio models use time-based pricing. Some catalogue rows describe additional distinctions, such as input plus output seconds, so read the full model rate card. EvoLink says failed generations are not billed. Per-key analytics and itemised records provide a way to check costs after a call.

Task handling and output limits

Model documentation can include constraints beyond the common API format. For example, the captured image-generation reference describes asynchronous processing using a returned task ID. It says generated image links remain valid for 24 hours.

An application using that documented endpoint therefore needs to track task completion and save its results within the stated window. The same reference includes file, mask, and format requirements. Check those details for the exact endpoint you intend to use; the platform's broad model access does not imply identical capabilities or response behaviour across models.