Open weights, one field away, cheaper past 2K tokens.

Same key, same base URL. Darkbloom serves quantized open-weight models on Apple Silicon at a fraction of the default rates, with a floor on every request.

Base URLhttps://api.minirouter.sh/v1

Get a key — no account

Models on Darkbloom

4 only here

Input rate, best case

3.6× lower

Google Gemma 4 26B A4B

Floor per request

$0.000105

whenever tokens cost less

Precision

2-bit / 4-bit / 8-bit

named per model

How it works

One line in the request. One header back.

the request · chat completions or the Responses API

 curl https://api.minirouter.sh/v1/chat/completions \   -H "Authorization: Bearer $MINIROUTER_KEY" \   -H "Content-Type: application/json" \   -d '{     "model": "openai/gpt-oss-20b",     "messages": [{"role": "user", "content": "ping"}],+added line:     "providerOptions": {"gateway": {"only": ["darkbloom"]}}   }'

the response

 HTTP/2 200+added line: x-minirouter-provider: openai/gpt-oss-20b@darkbloom x-minirouter-cost-usd: 0.000105 x-minirouter-request-id: req_…

Routing

Pinned means pinned. Unpinned means unchanged.

Rates

Per million tokens, side by side.

After markup, from the live catalog. The default route is what an unpinned request pays.

6 models with a Darkbloom route, current catalog rates
Model	Input, $ per 1M tokens	Output, $ per 1M tokens
Qwen3 VL 30B A3B Instruct4-bit on Darkbloom	$0.095 — Darkbloom only	$0.42 — Darkbloom only
Qwen 3.5 9B4-bit on Darkbloom	$0.084 — Darkbloom only	$0.137 — Darkbloom only
Qwen 3.6 35B A3B4-bit on Darkbloom	$0.053 — Darkbloom only	$0.735 — Darkbloom only
Google Gemma 4 26B A4B4-bit on Darkbloom	$0.044 $0.158 3.6× lower on Darkbloom	$0.231 $0.63 2.7× lower on Darkbloom
GPT OSS 20B8-bit on Darkbloom	$0.021 $0.032 1.5× lower on Darkbloom	$0.105 $0.147 1.4× lower on Darkbloom
Bonsai 2 27B2-bit on Darkbloom	$0.079 — Darkbloom only	$0.525 — Darkbloom only

DarkbloomDefault routeOne scale across both columns.

The floor

Cheaper past 2K input tokens.

GPT OSS 20B, reply held at 250 tokens. Below the crossing the default route wins; above it Darkbloom does, and the gap widens.

Cost per request on GPT OSS 20B, 250 output tokens, by prompt size, as a table
Input tokens	Darkbloom	Default route	Cheaper
0	$0.000105	$0.000037	Default
250	$0.000105	$0.000045	Default
500	$0.000105	$0.000053	Default
1K	$0.000105	$0.000068	Default
2K	$0.000105	$0.000100	Default
4K	$0.000110	$0.000163	Darkbloom
8K	$0.000194	$0.000289	Darkbloom

Four workloads on GPT OSS 20B

Health check

A ping.

+7592%

20 in · 5 out

Darkbloom: $0.000105floor
Default: $0.000001

Chat turn

Short history, short reply.

+156%

600 in · 150 out

Darkbloom: $0.000105floor
Default: $0.000041

Agent turn

System prompt, tools, history.

−32%

4K in · 250 out

Darkbloom: $0.000110
Default: $0.000163

Document pass

A long file in, a summary out.

−33%

30K in · 500 out

Darkbloom: $0.000682
Default: $0.0010

Pin the real work; leave health checks and empty pings on the default route.

Where to go from here.

Darkbloom provider pageModel list and disclosures.Routing and failoverHow the default route picks a host.Agent spend limitsCap an overnight agent.The full catalogEvery model and route, current rates.