API host · catalogue checked 27 Oct 2026

Azure: models, prices and uptime

Azure serves 13 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 served13of 78 we test

Cheapest host for5of 13 shared models

Median premium+100%over the cheapest host

Uptime90.9%last 30 min, average

Most accurate models on Azure

Verification score of each model this host serves

1Claude Opus 5.5Anthropic85.0%
2GPT-5.3-CodexOpenAI82.5%
3Claude Fable 5.1Anthropic81.2%
4Claude Sonnet 5.5Anthropic78.8%
5GPT-6 AstraOpenAI76.2%
6GPT-6.1 Terra ProOpenAI76.2%
7GPT-6.1 SolOpenAI76.2%
8GPT-5.6 TerraOpenAI75.0%
9GPT-6.1 TerraOpenAI73.8%
10GPT-6 SolOpenAI73.8%
11GPT-5.4OpenAI72.5%
12GPT-6 LunaOpenAI72.5%
13Claude Haiku 5.5Anthropic68.8%

Models on Azure

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
Claude Opus 5.5	85.0%	$4	$20	Cheapest	5	1M	128k	—	100%
GPT-5.3-Codex	82.5%	$1.75	$14	Cheapest	2	400k	128k	—	—
Claude Fable 5.1	81.2%	$10	$50	Cheapest	4	1M	128k	—	—
Claude Sonnet 5.5	78.8%	$2	$10	Cheapest	5	1M	128k	—	100%
GPT-6 Astra	76.2%	$10	$50	+100%	3	1.1M	128k	—	100%
GPT-6.1 Terra Pro	76.2%	$1.2	$6	+140%	2	1.1M	128k	—	100%
GPT-6.1 Sol	76.2%	$2	$10	+100%	3	1.1M	128k	—	0.0%
GPT-5.6 Terra	75.0%	$2	$12	+100%	3	1.1M	128k	—	100%
GPT-6.1 Terra	73.8%	$1.2	$6	+140%	2	1.1M	128k	—	100%
GPT-6 Sol	73.8%	$2	$10	+100%	3	1.1M	128k	—	100%
GPT-5.4	72.5%	$2.5	$15	+100%	3	1.1M	128k	—	100%
GPT-6 Luna	72.5%	$0.1	$0.5	+100%	3	1.1M	128k	—	100%
Claude Haiku 5.5	68.8%	$0.1	$0.5	Cheapest	5	1M	128k	—	100%

All API hosts →

Azure: frequently asked questions

Which AI models does Azure serve?: Of the models we benchmark, Azure serves 13 through OpenRouter: Claude Opus 5.5, GPT-5.3-Codex, Claude Fable 5.1, Claude Sonnet 5.5, GPT-6 Astra, GPT-6.1 Terra Pro, GPT-6.1 Sol, GPT-5.6 Terra, GPT-6.1 Terra, GPT-6 Sol, GPT-5.4, GPT-6 Luna, Claude Haiku 5.5.
What is the most accurate model on Azure?: Claude Opus 5.5 by Anthropic, with a verification score of 85.0% (68 of 80 questions correct) in our tests.
Is Azure the cheapest host?: Azure has the lowest list price for 5 of the 13 models it serves that other hosts also serve (Claude Opus 5.5, GPT-5.3-Codex, Claude Fable 5.1, Claude Sonnet 5.5, Claude Haiku 5.5). Its median markup over the cheapest host is +100%.
How reliable is Azure?: Averaged over the models it serves, Azure reported 90.9% 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

Claude Opus 5.5

85.0%

$20

Cheapest

128k

—

100%

GPT-5.3-Codex

82.5%

$1.75

$14

Cheapest

400k

128k

—

Claude Fable 5.1

81.2%

$10

$50

Cheapest

128k

—

Claude Sonnet 5.5

78.8%

$10

Cheapest

128k

—

100%

GPT-6 Astra

76.2%

$10

$50

+100%

1.1M

128k

—

100%

GPT-6.1 Terra Pro

76.2%

$1.2

+140%

1.1M

128k

—

100%

GPT-6.1 Sol

76.2%

$10

+100%

1.1M

128k

—

0.0%

GPT-5.6 Terra

75.0%

$12

+100%

1.1M

128k

—

100%

GPT-6.1 Terra

73.8%

$1.2

+140%

1.1M

128k

—

100%

GPT-6 Sol

73.8%

$10

+100%

1.1M

128k

—

100%

GPT-5.4

72.5%

$2.5

$15

+100%

1.1M

128k

—

100%

GPT-6 Luna

72.5%

$0.1

$0.5

+100%

1.1M

128k

—

100%

Claude Haiku 5.5

68.8%

$0.1

$0.5

Cheapest

128k

—

100%

Azure: frequently asked questions

Which AI models does Azure serve?

Of the models we benchmark, Azure serves 13 through OpenRouter: Claude Opus 5.5, GPT-5.3-Codex, Claude Fable 5.1, Claude Sonnet 5.5, GPT-6 Astra, GPT-6.1 Terra Pro, GPT-6.1 Sol, GPT-5.6 Terra, GPT-6.1 Terra, GPT-6 Sol, GPT-5.4, GPT-6 Luna, Claude Haiku 5.5.

What is the most accurate model on Azure?

Claude Opus 5.5 by Anthropic, with a verification score of 85.0% (68 of 80 questions correct) in our tests.

Is Azure the cheapest host?

Azure has the lowest list price for 5 of the 13 models it serves that other hosts also serve (Claude Opus 5.5, GPT-5.3-Codex, Claude Fable 5.1, Claude Sonnet 5.5, Claude Haiku 5.5). Its median markup over the cheapest host is +100%.

How reliable is Azure?

Averaged over the models it serves, Azure reported 90.9% uptime in the 30 minutes before OpenRouter's catalogue was read for our run.