Use Darkbloom as the AI model behind your tools
Updated
Short answer: Darkbloom is an OpenAI-compatible API, so any tool that lets you set a base URL and a key can use it. The base URL is https://api.darkbloom.dev/v1. Darkbloom runs open-weight models on Apple Silicon Macs, charges per token from a prepaid balance, and calls itself a public alpha. These guides give the exact settings for popular tools, each checked against that tool’s own docs.
Setup guides
How to use Darkbloom with Aider
Set up Aider with Darkbloom’s OpenAI-compatible API: base URL, API key, model ID, current models and prices, and fixes for the errors you are most likely to hit.
How to use Darkbloom with Cline
Set up Cline (VS Code or the cline command line) with Darkbloom’s OpenAI-compatible API: base URL, key, saving the model, cost, and the fix for the gpt-4o 403 error.
How to use Darkbloom with Continue
Set up Continue (VS Code, JetBrains or the cn command line) with Darkbloom’s OpenAI-compatible API: config.yaml, the key in a .env file, model ID, cost and fixes.
How to use Darkbloom with LiteLLM
Set up LiteLLM with Darkbloom’s OpenAI-compatible API, in Python or as a proxy: base URL, key, model ID, config.yaml, cost tracking and fixes for common errors.
How to use Darkbloom with Open WebUI
Set up Open WebUI with Darkbloom’s OpenAI-compatible API: edit the OpenAI connection, pick a model, see what one chat message costs, and fix the common errors.
How to use Darkbloom with n8n
Set up n8n with Darkbloom’s OpenAI-compatible API: the credential’s Base URL, the OpenAI Chat Model node, model choice, cost, and the fix for the gpt-5-mini 403 error.
The Darkbloom API at a glance
| Item | Value |
|---|---|
| Base URL | https://api.darkbloom.dev/v1 |
| Authentication | Authorization: Bearer <your key>. Create keys in the Darkbloom console. |
| Endpoints | /v1/chat/completions (OpenAI style), /v1/completions, /v1/messages (Anthropic style), /v1/models |
| Billing | Prepaid credits, charged per token, with a minimum of $0.0001 per request. Minimum deposit $0.50. Requests return 403 at a zero balance (Darkbloom docs). |
| Models | Open-weight models served by Macs that come and go. GET /v1/models lists what is online now. |
| Status | Public alpha. Darkbloom’s docs say not to use it in production applications. |
Which models are online, and what do they cost?
On October 23, 2026, GET /v1/models (which needs a key) listed these models. Prices are US dollars per million tokens and match the public GET /v1/pricing, which also names models that are not online. Model IDs change as Macs come and go, so call GET /v1/models with your key for the live list.
| Model ID | Context window, tokens | Input, $ per million tokens | Output, $ per million tokens | Features Darkbloom lists |
|---|---|---|---|---|
gpt-oss-20b | 131,072 | $0.018 | $0.09 | none listed |
nvidia-nemotron-3.5-lightning | 262,144 | $0.025 | $0.12 | tools, reasoning |
EigenLabs/Ternary-Bonsai-2-9B-MLX-2bit | 262,144 | $0.08 | $0.13 | tools, reasoning |
Qwen3.5-9B | 262,144 | $0.08 | $0.13 | tools, image input |
glm-5.3-flash | 262,144 | $0.065 | $0.18 | tools, reasoning, image input |
gemma-4-26b | 131,072 | $0.042 | $0.22 | tools, reasoning, json mode, image input |
mimo-v2.6-flash-mopd | 1,048,576 | $0.10 | $0.28 | tools, reasoning |
nvidia-nemotron-3.5-super | 262,144 | $0.075 | $0.35 | tools, reasoning, json mode, image input |
ternary-bonsai-2-27b | 262,144 | $0.075 | $0.50 | tools, reasoning, image input |
qwen3.6-35b-a3b-vl-mtp-mxfp8 | 262,144 | $0.05 | $0.70 | tools, image input |
qwen3.5-35b-a3b | 262,144 | $0.065 | $0.75 | tools, image input |
EigenLabs/Qwen3.8-27B-4bit-mtp | 262,144 | $0.035 | $2.20 | tools, image input |
Questions
Is the Darkbloom API compatible with OpenAI?
Yes. It accepts the OpenAI Chat Completions format at https://api.darkbloom.dev/v1/chat/completions, with streaming, and it also accepts the Anthropic Messages format at /v1/messages. For most tools the only changes are the base URL, the key and the model ID.
What does the Darkbloom API cost?
It is priced per million tokens, and list prices vary by model. On October 23, 2026, output prices ran from $0.09 to $2.20 per million tokens, and every request costs at least $0.0001. Live prices are public at GET /v1/pricing.
Which models can I use?
Only models that a Mac is serving right now. The list changes, so call GET /v1/models with your key. Darkbloom’s own docs list model IDs that are no longer online, so do not copy them. A key can also be limited to some models, and any other model gets a 403.
Is it safe to send private code to Darkbloom?
Not yet. Darkbloom is a public alpha. It says prompts are encrypted to a verified provider Mac, and an independent review (Wavect, September 2026) reads its documentation as saying that the coordinator and the provider both handle the prompt as plaintext inside protected boundaries. Use it for non-sensitive work.
About these guides
BloomGauge is a free Mac app for people who run Darkbloom providers, so we benefit when Darkbloom has more paying demand. We are independent of Darkbloom and not endorsed by it. Our Darkbloom sign-in links carry our referral code, BLOOMGAUGE. If you sign in to Darkbloom after opening one, Darkbloom attaches the code to your account and pays us a share of your eligible token spend. Darkbloom funds that reward, so it does not change your price.