API host · catalogue checked 27 Oct 2026

Darkbloom: models, prices and uptime

Darkbloom serves 8 of the 78 models we benchmark through OpenRouter. Prices are OpenRouter's list prices per million tokens. We did not measure speed per host.

Models served8of 78 we test

Cheapest host for6of 8 shared models

Median premiumCheapestover the cheapest host

Uptime100%last 30 min, average

Most accurate models on Darkbloom

Verification score of each model this host serves

1GLM 5.3 FlashZ.ai67.5%
2Qwen3.8 27BQwen63.8%
3MiMo-V2.6-FlashXiaomi62.5%
4gpt-oss-20bOpenAI51.2%
5Nemotron 3.5 LightningNVIDIA43.8%
6Nemotron 3.5 Super VLNVIDIA40.0%
7Qwen3.6 35B A3BQwen37.5%
8Gemma 4 26B A4BGoogle15.0%

Models on Darkbloom

Premium: this host's blended price (3 input : 1 output) against the cheapest host for the same model

Model	Score	Input / 1M	Output / 1M	vs cheapest host	Hosts	Context	Max output	Quantization	Uptime
GLM 5.3 Flash	67.5%	$0.065	$0.18	+100%	34	262k	66k	fp4	100%
Qwen3.8 27B	63.8%	$0.035	$2.2	+239%	21	262k	33k	fp4	100%
MiMo-V2.6-Flash	62.5%	$0.1	$0.28	Cheapest	10	1M	131k	fp4	100%
gpt-oss-20b	51.2%	$0.018	$0.09	Cheapest	10	131k	33k	fp8	100%
Nemotron 3.5 Lightning	43.8%	$0.025	$0.12	Cheapest	4	262k	33k	int4	100%
Nemotron 3.5 Super VL	40.0%	$0.06	$0.35	Cheapest	9	262k	33k	nvfp4	100%
Qwen3.6 35B A3B	37.5%	$0.05	$0.7	Cheapest	10	262k	33k	fp4	100%
Gemma 4 26B A4B	15.0%	$0.042	$0.22	Cheapest	15	131k	33k	—	100%

All API hosts →

Darkbloom: frequently asked questions

Which AI models does Darkbloom serve?: Of the models we benchmark, Darkbloom serves 8 through OpenRouter: GLM 5.3 Flash, Qwen3.8 27B, MiMo-V2.6-Flash, gpt-oss-20b, Nemotron 3.5 Lightning, Nemotron 3.5 Super VL, Qwen3.6 35B A3B, Gemma 4 26B A4B.
What is the most accurate model on Darkbloom?: GLM 5.3 Flash by Z.ai, with a verification score of 67.5% (54 of 80 questions correct) in our tests.
Is Darkbloom the cheapest host?: Darkbloom has the lowest list price for 6 of the 8 models it serves that other hosts also serve (MiMo-V2.6-Flash, gpt-oss-20b, Nemotron 3.5 Lightning, Nemotron 3.5 Super VL, Qwen3.6 35B A3B, Gemma 4 26B A4B). Its median markup over the cheapest host is Cheapest.
How reliable is Darkbloom?: Averaged over the models it serves, Darkbloom reported 100% uptime in the 30 minutes before OpenRouter's catalogue was read for our run.

Model

Score

Input / 1M

Output / 1M

vs cheapest host

Hosts

Context

Max output

Quantization

Uptime

GLM 5.3 Flash

67.5%

$0.065

$0.18

+100%

262k

66k

fp4

100%

Qwen3.8 27B

63.8%

$0.035

$2.2

+239%

262k

33k

fp4

100%

MiMo-V2.6-Flash

62.5%

$0.1

$0.28

Cheapest

131k

fp4

100%

gpt-oss-20b

51.2%

$0.018

$0.09

Cheapest

131k

33k

fp8

100%

Nemotron 3.5 Lightning

43.8%

$0.025

$0.12

Cheapest

262k

33k

int4

100%

Nemotron 3.5 Super VL

40.0%

$0.06

$0.35

Cheapest

262k

33k

nvfp4

100%

Qwen3.6 35B A3B

37.5%

$0.05

$0.7

Cheapest

262k

33k

fp4

100%

Gemma 4 26B A4B

15.0%

$0.042

$0.22

Cheapest

131k

33k

—

100%

Darkbloom: frequently asked questions

Which AI models does Darkbloom serve?

Of the models we benchmark, Darkbloom serves 8 through OpenRouter: GLM 5.3 Flash, Qwen3.8 27B, MiMo-V2.6-Flash, gpt-oss-20b, Nemotron 3.5 Lightning, Nemotron 3.5 Super VL, Qwen3.6 35B A3B, Gemma 4 26B A4B.

What is the most accurate model on Darkbloom?

GLM 5.3 Flash by Z.ai, with a verification score of 67.5% (54 of 80 questions correct) in our tests.

Is Darkbloom the cheapest host?

Darkbloom has the lowest list price for 6 of the 8 models it serves that other hosts also serve (MiMo-V2.6-Flash, gpt-oss-20b, Nemotron 3.5 Lightning, Nemotron 3.5 Super VL, Qwen3.6 35B A3B, Gemma 4 26B A4B). Its median markup over the cheapest host is Cheapest.

How reliable is Darkbloom?

Averaged over the models it serves, Darkbloom reported 100% uptime in the 30 minutes before OpenRouter's catalogue was read for our run.