Cline pricing with Vana API: GPT-5.6 Sol at $0.3472

Agents at the terminalEN2026-11-06

§ One Vana API address, three protocols

Your tools can share one Vana API address: https://api.vanaapi.com. That address answers API requests for the three protocol styles Vana API supports, so you do not need a separate endpoint for every editor, terminal and script in your workflow. The point is simple to hold onto: one address, three protocols, minimal setup.

The OpenAI format runs on /v1/chat/completions. It is the format used by Cursor, Cline, Roo Code, OpenCode, Cherry Studio, any OpenAI SDK and cURL. GPT models also answer on /v1/responses, which is what Codex uses. That means a tool built around the OpenAI format can point at the same Vana API address as another tool with a different model preference, without changing where requests go.

The other two protocols sit on the same address. The Anthropic Messages protocol (/v1/messages) is spoken by official Claude Code or Claude Desktop, version 2.1.285 or later, which handle those requests themselves. The Gemini native protocol (/v1beta/models/...) is used by Gemini CLI and is not OpenAI-compatible. Keeping these routes distinct matters, because a client that expects one protocol should not be pointed at another.

One detail worth separating early: https://api.vanaapi.com answers API requests only. Your dashboard, where keys, balance and the usage log live, is on the main site at https://vanaapi.com. So the setup is one request address plus one place to manage the account, and the address itself does not change when you move between tools.

§ Connecting Cline and other tools

Cline uses the OpenAI format on /v1/chat/completions, so it pairs with GPT models and with the other models Vana API lists in that format: Grok, DeepSeek, GLM, Kimi, Qwen, MiniMax and MiMo. Cursor, Roo Code, OpenCode and Cherry Studio work the same way. Any OpenAI SDK or cURL command can reach the same endpoint. This is what makes the shared-address idea practical: the connection details stay fixed while the model changes.

Minimal setup here means fewer moving parts rather than fewer steps. You create a key in the Vana API dashboard, add the base address https://api.vanaapi.com to the client, and choose an OpenAI-format model. There is no monthly fee to add at this stage either, because Vana API works on a prepaid balance charged per real token.

Because every call is listed in the usage log in the dashboard, you can check what a tool actually spent. If Cline, Cursor and a script all point at the same Vana API address, the log still keeps them in one place. You can read the balance, review the calls and see the tokens charged without assembling separate reports from each client.

For example, say you run Cline for coding during the week and a cURL script for a small internal task. Both can send OpenAI-format requests to https://api.vanaapi.com. The script does not need a different relay address, and Cline does not need a separate one either. That is the whole design: one address, three protocols, little to set up.

§ What Cline-style requests cost

Prices below are per million tokens, shown as input / output. GPT-5.6 Sol, model id gpt-5.6-sol, is listed at $0.3472 / $1.81 on Vana API, against an official list price of $4 / $20. GPT-6 Astra, model id gpt-6-astra, is $0.6944 / $3.54 on Vana API, against $10 / $50. GPT-6.1 Sol, model id gpt-6.1-sol, is $0.1389 / $0.8333 on Vana API, against $2 / $10.

The rest of the OpenAI-format list is priced the same way: Grok 4.7 (grok-4.7) at $0.4861 / $1.53, DeepSeek V4 Flash (deepseek-v4-flash) at $0.2778 / $1.04, and DeepSeek V4 Pro (deepseek-v4-pro) at $1.11 / $3.33.

GLM-5.3 (glm-5.3): $0.5556 / $1.67
Kimi K3 (kimi-k3): $1.04 / $5.21
Qwen3.7 Plus (qwen3.7-plus): $0.1278 / $0.4167
Qwen3.8 Max (qwen3.8-max): $0.6944 / $2.08
MiniMax M3 (minimax-M3): $0.1389 / $0.5556
MiMo V2.6 Pro (mimo-V2.6-Pro): $0.2083 / $0.3472

Since Cline can reach any of these through the OpenAI format, the useful question is not which tool to keep but which model fits the job and the budget. The per-token rates make that comparison direct, and the usage log shows what you actually paid for after the calls are made.

§ Prepaid balance and what you can see

Vana API billing is a prepaid balance charged per real token, with no monthly fee. That keeps the cost tied to use rather than to a subscription cycle. If Cline is busy one week and quiet the next, the balance follows the activity. There is no plan to renew just to keep the same address available.

Keys are created in the dashboard, and every call is listed in the usage log there. Those two facts together give you a simple routine: create a key, point your client at https://api.vanaapi.com, then check the log to see the calls and the balance. You do not need to guess whether a client is using a key correctly, because the entries are already recorded on the same account.

There is also a small starting credit on sign-up: $0.10 of credit. It is a modest amount rather than a headline, and it is enough to make a first request and confirm that the address, the key and the model id line up before you commit a larger balance.

If you invite others, the referral arrangement pays 14% of what the people you invite pay, and 5% one level further. Earnings can be withdrawn from $13.89. Refunds are available on unused balance within 3 days of the top-up, minus what was used and the payment processor's fee.

§ A monthly plan for steady GPT use

If your GPT usage is regular and predictable, Vana API also offers a monthly plan: GPT up to $2,750 / 30 days: $103.61 for 30 days, at most $820 a week, GPT models only. The plan is scoped to GPT models, so it suits a workflow built around those ids rather than one that mixes in DeepSeek, GLM, Kimi, Qwen, MiniMax, MiMo or Grok requests on the same schedule.

The shape of the offer is straightforward. It covers 30 days, it caps weekly use at $820, and it is priced at $103.61 for 30 days. For a team or an individual whose coding work in Cline runs on GPT models every day, that gives a predictable ceiling to plan around instead of a variable prepaid balance.

Either way, the tooling stays the same. Cline, Cursor, Roo Code, OpenCode, Cherry Studio, an OpenAI SDK or cURL still point at https://api.vanaapi.com, and the dashboard still holds your keys, balance and usage log. The choice between prepaid pay-as-you-go and the monthly plan is about how you want to fund the calls, not about where they go.

One address, three protocols and minimal setup is a small idea, but it removes a common source of friction: every client asking for its own endpoint. With Vana API, the request address is fixed, the protocol follows the tool, and the model id follows your choice. You set the key once in the dashboard, connect the tools you already use, and read the log when you want to see what the calls cost.