Integration guide

Cline on Seedrise

To use Cline with Seedrise, set API Provider to OpenAI Compatible, enter the base URL https://api.seedrise.ai/v1 and your Seedrise key, then pick a model ID such as claude-sonnet-5-5.

Updated 2026-11-02

Get a rate card →Jump to setup

Cline settings at a glance

Cline field	Value for Seedrise
API Provider	OpenAI Compatible
Base URL	`https://api.seedrise.ai/v1`
OpenAI Compatible API Key	Your Seedrise key
Model ID	Any ID on your key, for example `claude-sonnet-5-5`, `gpt-6.1-sol` or `deepseek-v3.2`
Context Window Size	The model's own window, from the table below

Field names checked against Cline's settings panel on 2026-11-02.

Why run Cline through Seedrise

Every model on one key

Claude, GPT, Gemini, DeepSeek, Kimi and more behind one base URL. Cline reads the model list from your key, so switching models is a dropdown, not a new provider.

Hard limits on every key

Budget, model allow-list, IP allow-list and expiry on each key. A runaway task stops at the budget you set instead of running on.

One invoice, Australian regions

Usage by key and model in the console, one monthly invoice in USD or AUD. Models marked AU can be pinned to Sydney or Melbourne on request.

Setup in five minutes

Get a Seedrise key. Request a rate card, receive your console login, and create a key for Cline in the console. Add the models you want to its allow-list and set a monthly budget.
Choose the provider. Click the gear icon in the Cline panel and set API Provider to OpenAI Compatible.
Enter the base URL and key. Set Base URL to https://api.seedrise.ai/v1, keeping /v1 at the end; Cline uses it exactly as entered. Put your Seedrise key in OpenAI Compatible API Key.
Base URL
```
https://api.seedrise.ai/v1
```
Pick a model. Cline fills the Model ID list from GET /v1/models, so it shows the models on your key. Choose one such as claude-sonnet-5-5, or pick Use custom model ID… and type it. No vendor prefix.
Match the model configuration. Open Model Configuration and set Context Window Size to the model's window from the table below; Cline assumes 128,000 for a custom model. Untick Supports Images for text-only models such as DeepSeek V3.2. Then start a task.

Model settings for Cline

Values for Cline's Model Configuration panel, typed without separators. Specs as at 1 November 2026; every model and its list price is in the model library.

Model	Model ID	Context Window Size	Max Output Tokens	Supports Images
Claude Sonnet 5.5	`claude-sonnet-5-5`	`1000000`	`128000`	Yes
Claude Opus 5.5	`claude-opus-5-5`	`1000000`	`128000`	Yes
GPT-6.1 Sol	`gpt-6.1-sol`	`1050000`	`128000`	Yes
DeepSeek V3.2	`deepseek-v3.2`	`164000`	`8000`	No
Kimi K2.5	`kimi-k2.5`	`256000`	`16000`	Yes

Claude through Cline's Anthropic provider

For Claude-only work you can use Cline's Anthropic provider instead. It speaks Anthropic Messages, which Seedrise serves at the root URL, and Cline supports Claude prompt caching on it, so repeated context in long sessions is billed at the cache-read rate.

Cline field	Value for Seedrise
API Provider	Anthropic
Anthropic API Key	Your Seedrise key
Use custom base URL	Tick it and enter `https://api.seedrise.ai`, with no `/v1`
Model	`claude-opus-5-5`, `claude-sonnet-5-5` or `claude-haiku-5-5`, as listed in Cline

The Claude IDs in Cline's model list are the same IDs Seedrise uses, so no custom ID is needed.
This provider reaches Claude models only. Keep OpenAI Compatible for GPT, DeepSeek, Kimi and the rest of your key.

Check it is working

The Model ID list shows the models on your key. If it stays empty, the base URL or key is wrong; check both with the request below.
Your first task runs as usual, and usage appears in the Seedrise console within a minute, by key and model.
Cline's cost display uses the Input Price / 1M tokens and Output Price / 1M tokens fields, which start at 0. Enter your rate-card prices there if you want it to match your invoice.

bash

# lists the model IDs your key can call
curl https://api.seedrise.ai/v1/models \
  -H "Authorization: Bearer sk-..."

Known errors

Status	Meaning	Fix
`400`	Bad request, usually a wrong model ID or malformed body	Check the ID against `GET /v1/models`
`401`	Key missing, disabled or expired	Check the key in the console
`403`	Model not on this key, or IP not in the allow-list	Adjust the key's model list or allow-list
`404`	Wrong base URL path, or an endpoint Seedrise does not serve	OpenAI Compatible needs `https://api.seedrise.ai/v1`; the Anthropic provider takes `https://api.seedrise.ai`
`429`	Rate limit or budget reached	Back off and retry; raise the limit in the console or ask us
`5xx`	Upstream or gateway error	Retry with backoff; the gateway fails over across routes automatically

Still stuck? Email support@seedrise.ai with the time of the request, the model ID and the key name.

For teams

Give each developer their own key rather than sharing one.

Per-developer keys show usage by person in the console, and a leaked key can be revoked without touching anyone else.
A shared team key is simpler to hand out, but spend cannot be split by person and every rotation reaches the whole team.
Set a monthly budget on every key. When a key reaches its budget, requests return 429 and the agent stops instead of running on.
Limit team keys to the models you approve, and add an IP allow-list for your office or VPN egress.
If your organisation manages Cline centrally with remote configuration, the base URL can be set and locked there; locked fields show a lock icon.

Cline questions

Which Cline provider should I choose for Seedrise? +

OpenAI Compatible, with the base URL https://api.seedrise.ai/v1. It reaches every model on your key. For Claude-only use, the Anthropic provider with the custom base URL https://api.seedrise.ai also works and supports Claude prompt caching.

Why is the Model ID list empty? +

Cline loads the list from GET /v1/models on your base URL. An empty list usually means the base URL is missing /v1, the key is wrong or disabled, or no models are on the key yet. You can still type a model ID by hand.

What context window should I set in Cline? +

The model's own window: 1000000 for Claude Opus, Sonnet and Haiku 5.5, 1050000 for GPT-6.1 Sol, 164000 for DeepSeek V3.2 and 256000 for Kimi K2.5, as at 1 November 2026. Otherwise Cline manages the conversation as if every custom model had a 128,000-token window. Details per model are in the model library.

Can I keep inference in Australia? +

Yes. Models marked AU, including Claude Opus, Sonnet and Haiku 5.5, DeepSeek V3.2 and Kimi K2.5, can be pinned to Sydney or Melbourne on request. Pinning is set per key, so nothing changes in Cline.

How is Cline usage billed? +

Per token, at the rates on your rate card, with input, output and cached input itemised. One monthly invoice in USD or AUD covers every model and key, and there is no seat fee per Cline user. See plans and pricing.

Ready to run Cline on Seedrise?

Per-model rates in writing within one business day, then a console login and a trial key.

Get a rate card →

Also: Claude Code setup · LiteLLM setup · full docs · plans and pricing · security and data

Seedrise is an independent provider and is not affiliated with or endorsed by Anthropic. Claude is a trademark of Anthropic, PBC.