Cline with an OpenAI-compatible API: connect Heabsy
In Cline choose the API provider "OpenAI Compatible", set the base URL to https://api.heabsy.com/v1, paste your Heabsy key and enter a model ID such as qwen38. Then fill in Model Configuration — context window and max output — so Cline manages the conversation with the right limits.
API Provider: OpenAI Compatible
Base URL: https://api.heabsy.com/v1
API Key: <your Heabsy key>
Model ID: qwen38https://api.heabsy.com/v1 · Anthropic base URL https://api.heabsy.comStep by step
- Open Cline in VS Code and click the gear icon.
- API Provider: OpenAI Compatible. Base URL:
https://api.heabsy.com/v1— without/chat/completions. - API Key: your Heabsy key. Model ID: the exact catalog ID, for example
qwen38ormimo-v26-flash. - Click Verify.
- Open Model Configuration and enter Context Window and Max Output Tokens from the table below, and enable Image Support for models that read images. Input and output prices are optional: Cline has no field for the cached-input price, so its cost counter treats cached input as free and shows less than you pay (Cline issue #13944). The Logs page of your Heabsy console shows what each request was billed.
Which model should I pick?
Cline drives everything through tool calls, so a model without tool calling will not work. All models in the table support it; our own run in the EEA, all except DeepSeek V4.1 Flash Uncensored, without storing prompts.
| Model | ID | Context | Max output | Runs | Input / 1M | Output / 1M |
|---|---|---|---|---|---|---|
| Qwen3.8 27B Fable | qwen38-fable | 512K | 32K | EEA | $0.10 | $0.61 |
| Qwen3.8 27B Queen | qwen38-queen | 512K | 32K | EEA | $0.10 | $0.61 |
| GLM 5.3 Flash Uncensored | glm53-flash-unc | 1M | 128K | EEA | $0.20 | $0.8001 |
| Qwen3.8 27B Obliterated | qwen38-obl | 512K | 32K | EEA | $0.21 | $1.47 |
| DeepSeek V4.1 Flash 1017 Uncensored | deepseek-v4.1-flash-1017-unc | 1M | 64K | outside EEA | $0.40 | $1.40 |
| DeepSeek V4.1 Flash Uncensored | deepseek-v4.1-flash-unc | 1M | 64K | outside EEA | $0.40 | $1.40 |
| MiMo V2.6 Flash Uncensored | mimo-v26-flash-unc | 1M | 64K | EEA | $0.4624 | $1.4999 |
| Qwen3.8 27B Uncensored | qwen38-unc | 512K | 230K | EEA | $0.4887 | $2.17 |
| MiMo V2.6 Flash | mimo-v26-flash | 1M | 64K | EEA | $0.50 | $1.50 |
| MiMo V2.6 Flash Abliterated | mimo-v26-flash-abl | 1M | 64K | EEA | $0.50 | $1.50 |
| Qwen3.8 27B Uncensored Fast by Heabsy | qwen38-unc-fast | 512K | 230K | EEA | $0.8299 | $2.62 |
| Qwen3.8 27B | qwen38 | 512K | 32K | EEA | $1.05 | $3.15 |
| CyberHeabsy | cyberheabsy | 1M | 64K | EEA | $1.50 | $4.50 |
All models above run on Heabsy hardware in the EEA, except DeepSeek V4.1 Flash Uncensored; prompts and completions are not stored. Thecatalog also lists 24 partner-hosted models — Claude, GPT-6, Kimi K3, DeepSeek V4 and others — with the same key and invoice, but without zero data retention and without an EEA guarantee.
Verify it works
Ask Cline to list the files of your workspace. If it calls the tool and shows the list, the connection and tool calling both work; the request also appears in the Logs page of your Heabsy console.
If the tool shows an error you cannot read, test the key on its own first — one short request outside the tool:
curl https://api.heabsy.com/v1/chat/completions \
-H "Authorization: Bearer $HEABSY_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"qwen38","max_tokens":16,"messages":[{"role":"user","content":"Say hi"}]}'Troubleshooting
| What you see | What to change |
|---|---|
| 404 Not Found | The base URL contains /chat/completions or lacks /v1. Use exactly https://api.heabsy.com/v1. |
| The model answers in text but never uses tools | Pick a model with tool calling from the table, and make sure Cline is on a current version — native tool calling for OpenAI Compatible arrived in early 2026. |
| Cline truncates or summarises too early | Model Configuration still has the default window. Enter the model's context window from the table. |
| 401 Unauthorized | The key field is empty or has a space at the end. Paste it again from the console. |
| Cline shows a lower cost than the console | For OpenAI Compatible models Cline counts cached input at $0, because the settings have no field for its price (Cline issue #13944, open since September 2026). In an agent session most of the input is cached, so the gap is large. The Logs page of your console shows the billed amount per request. |
| 403 with an HTML page ("Sorry, you have been blocked") | No key was sent at all: the key variable is empty or not set where the tool runs. Requests without a key are stopped before they reach the API; a wrong key gets a JSON 401 instead. |
Where do my prompts go?
From Cline straight to api.heabsy.com. On our own models the request is processed in the European Economic Area, except DeepSeek V4.1 Flash Uncensored and the text is not stored — we keep what billing needs: the model, the token counts and the time. Partner-hosted models are processed by third-party inference providers; each is marked in the catalog, and none of them carries the no-retention promise. The terms are in the data processing agreement, the invoice comes from an EU company.
FAQ
Does Cline work with Qwen3.8 27B?+
Yes. qwen38 supports tool calling, has a 524,288-token window and costs $1.05 / $3.15 per million input / output tokens; it runs on our own hardware in the EEA.
Does prompt caching work through OpenAI Compatible?+
Yes, on our own models it needs no setting: requests of one conversation land on the GPU that already holds its prefix, and cached input is billed at the cached price shown in the catalog. Cline's own cost counter has no cached price and shows that part as free; the Logs page of your console shows the real charge.
Coming from Roo Code?+
Roo Code shut down on 15 May 2026 and named Cline as a successor. The OpenAI Compatible settings are the same: base URL, key and model ID carry over unchanged.
Can I keep Cline's own models and add Heabsy?+
Cline uses one provider per configuration profile. Create a second profile for Heabsy and switch between them.
Longer walkthrough with costs: Cline on Qwen3.8 27B: setup, a measured task and what it cost
Also set up:Kilo Code · OpenCode · Claude Code
Cline is an open-source project; Heabsy is not affiliated with it.