Skip to content
← Back to blog

Use DeepSeek with Cline, Continue and Aider through one OpenAI-compatible API

By Reeva Team · October 28, 2026 · Updated November 17, 2026

DeepSeek v4.1 is one of the three models in the Reeva AI Models API. Cline, Continue and Aider can each use an OpenAI-compatible provider, so you can point them at Reeva and choose deepseek-v4.1-flash as the model. This post shows the settings for each tool. Plans and prices are on the AI Models API page.

What you need

  • A Reeva AI API key. Your first key is issued when the PayPal payment clears. You can create up to 10 keys, each with its own name and an optional expiry, and revoke any one of them on its own.
  • The base URL: https://api.reeva.me/v1
  • The model name: deepseek-v4.1-flash. The other two models are qwen3.6-35b-a3b (Qwen 3.6, for short, quick tasks; up to 32k tokens per request on every plan) and qwen-image (images).

Keep the key in an environment variable or in the tool’s own secret storage, not in a file you commit.

Cline

In the Cline settings panel (the gear icon), set these fields. The Cline docs name them as follows:

  • API Provider: “OpenAI Compatible”
  • Base URL: https://api.reeva.me/v1
  • API Key: your Reeva key
  • Model: deepseek-v4.1-flash, entered as the model ID

Source: Cline, OpenAI Compatible provider.

Continue

In Continue’s config file, add a model entry that uses the OpenAI provider with a custom apiBase. The Continue docs name these fields:

  • provider: openai
  • model: deepseek-v4.1-flash
  • apiBase: https://api.reeva.me/v1
  • apiKey: your Reeva key. Do not paste the key into a file you commit.

Source: Continue, OpenAI provider.

Aider

Aider reads two environment variables, and the model name takes an openai/ prefix:

export OPENAI_API_BASE=https://api.reeva.me/v1
export OPENAI_API_KEY=$REEVA_API_KEY
aider --model openai/deepseek-v4.1-flash

Source: Aider, OpenAI compatible APIs.

What your plan sets

  • Requests per minute: Mini 5, Starter 20, Pro 60, Scale 120. A request over the per-minute limit gets HTTP 429.
  • Requests at once: Mini and Starter 1, Pro 4, Scale 6. A request over this limit waits for one of your own requests to finish, and gets HTTP 429 only if no slot frees up within 100 seconds.
  • Request size: a request is refused with HTTP 400 and request_too_large if the prompt plus max_tokens is bigger than your plan allows: 32k tokens on Mini, 128k on Starter and Pro. Scale has no size cap of its own, only the 256k context of DeepSeek v4.1.
  • Tokens: there is no monthly token cap and no per-token billing. Fair use applies: heavy use slows requests down, and only extreme use within a single day is paused.

Source for all of the above: the AI Models API page, its plan table and its FAQ.

Check it works

Ask the tool for a short answer first, such as “Reply with OK.” If you get HTTP 401, the key is wrong or was revoked. If you get HTTP 429 at once, you are over the per-minute limit: wait a minute and retry.

Next steps

Choose your plan on the AI Models API page. For the full setup with code examples, read how to connect an OpenAI-compatible tool to Reeva AI.