Claude Haiku 5.5: providers compared

Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.

Current state

	State
Amazon Bedrockamazon-bedrock	Healthy	100%◆ (best)	100%	99.91%	99.91%	2.54s	9.68s	170 tps	1.0M	$0.100 / $0.500◆ (best)
Amazon Bedrockamazon-bedrock/us-east-1	Healthy	100%◆ (best)	100%	100%◆ (best)	100%◆ (best)	1.88s	3.46s	182 tps◆ (best)	1.0M	$0.110 / $0.550
Anthropicanthropic	Healthy	100%◆ (best)	100%	100%◆ (best)	100%◆ (best)	1.15s	11.0s	143 tps	1.0M	$0.100 / $0.500◆ (best)
Azureazure/global	Healthy	100%◆ (best)	100%	100%	100%	1.21s	21.5s	154 tps	1.0M	$0.100 / $0.500◆ (best)
Claude Platform on AWSclaude-on-aws	Healthy	100%◆ (best)	100%	99.98%	99.98%	2.42s	18.1s	162 tps	1.0M	$0.100 / $0.500◆ (best)
Googlegoogle-vertex/europe	Healthy	100%◆ (best)	100%	99.87%	99.87%	621ms	3.76s	105 tps	1.0M	$0.110 / $0.550
Googlegoogle-vertex/global	Healthy	100%◆ (best)	100%	100%◆ (best)	100%◆ (best)	907ms	7.96s	137 tps	1.0M	$0.100 / $0.500◆ (best)
Googlegoogle-vertex/us	Healthy	100%◆ (best)	100%	100%◆ (best)	100%◆ (best)	499ms◆ (best)	4.37s	87 tps	1.0M	$0.110 / $0.550
Amazon Bedrockamazon-bedrock/eu-west-1	No data	—	100%	100%◆ (best)	100%◆ (best)	831ms	12.0s	121 tps	1.0M	$0.110 / $0.550
Azureazure/us	No data	—	100%	100%◆ (best)	100%◆ (best)	720ms	3.26s◆ (best)	53 tps	1.0M	$0.110 / $0.550

Amazon Bedrock100%◆ (best)
Healthy24h 100%
7d
99.91%
30d
99.91%
p50
2.54s
p99
9.68s
Throughput
170 tps
Context
1.0M
$/Mtok
$0.100 / $0.500◆ (best)
amazon-bedrock
Amazon Bedrock100%◆ (best)
Healthy24h 100%
7d
100%◆ (best)
30d
100%◆ (best)
p50
1.88s
p99
3.46s
Throughput
182 tps◆ (best)
Context
1.0M
$/Mtok
$0.110 / $0.550
amazon-bedrock/us-east-1
Anthropic100%◆ (best)
Healthy24h 100%
7d
100%◆ (best)
30d
100%◆ (best)
p50
1.15s
p99
11.0s
Throughput
143 tps
Context
1.0M
$/Mtok
$0.100 / $0.500◆ (best)
anthropic
Azure100%◆ (best)
Healthy24h 100%
7d
100%
30d
100%
p50
1.21s
p99
21.5s
Throughput
154 tps
Context
1.0M
$/Mtok
$0.100 / $0.500◆ (best)
azure/global
Claude Platform on AWS100%◆ (best)
Healthy24h 100%
7d
99.98%
30d
99.98%
p50
2.42s
p99
18.1s
Throughput
162 tps
Context
1.0M
$/Mtok
$0.100 / $0.500◆ (best)
claude-on-aws
Google100%◆ (best)
Healthy24h 100%
7d
99.87%
30d
99.87%
p50
621ms
p99
3.76s
Throughput
105 tps
Context
1.0M
$/Mtok
$0.110 / $0.550
google-vertex/europe
Google100%◆ (best)
Healthy24h 100%
7d
100%◆ (best)
30d
100%◆ (best)
p50
907ms
p99
7.96s
Throughput
137 tps
Context
1.0M
$/Mtok
$0.100 / $0.500◆ (best)
google-vertex/global
Google100%◆ (best)
Healthy24h 100%
7d
100%◆ (best)
30d
100%◆ (best)
p50
499ms◆ (best)
p99
4.37s
Throughput
87 tps
Context
1.0M
$/Mtok
$0.110 / $0.550
google-vertex/us
Amazon Bedrock—
No data24h 100%
7d
100%◆ (best)
30d
100%◆ (best)
p50
831ms
p99
12.0s
Throughput
121 tps
Context
1.0M
$/Mtok
$0.110 / $0.550
amazon-bedrock/eu-west-1
Azure—
No data24h 100%
7d
100%◆ (best)
30d
100%◆ (best)
p50
720ms
p99
3.26s◆ (best)
Throughput
53 tps
Context
1.0M
$/Mtok
$0.110 / $0.550
azure/us

Measured 3m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.

Price vs latency

One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag

History

Availability

Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.

Time to first token

Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window, averaged over the hour.

Throughput

Median tokens per second.