Start typing to search.
Ranks providers by their recent typical output speed in tokens per second. Providers without enough recent data remain visible after ranked rows.
Google Vertex currently ranks first at 78.0 tok/s via google-vertex/europe.
| Rank | Provider / route | Output speed | Speed measurements | Endpoint details |
|---|---|---|---|---|
| #1 | Provider / route Google Vertex google-vertex/europe | Output speed78.0 tok/s | Speed measurements
| Endpoint details
|
| #2 | Provider / route Amazon Bedrock amazon-bedrock/eu-west-1 | Output speed71.5 tok/s | Speed measurements
| Endpoint details
|
| #3 | Provider / route Google Vertex google-vertex/global | Output speed67.0 tok/s | Speed measurements
| Endpoint details
|
| #4 | Provider / route Amazon Bedrock amazon-bedrock/global | Output speed65.0 tok/s | Speed measurements
| Endpoint details
|
| #5 | Provider / route Azure azure/us-east-2 | Output speed58.0 tok/s | Speed measurements
| Endpoint details
|
| #6 | Provider / route Anthropic anthropic | Output speed57.0 tok/s | Speed measurements
| Endpoint details
|
Endpoint data fetched .
Claude Haiku 4.5 endpoint guide
6 of 6 Claude Haiku 4.5 endpoints currently have the published data required for this ranking. Google Vertex leads at 78.0 tok/s via google-vertex/europe. The table keeps unranked routes visible so missing measurements do not look like missing provider availability.
The output-speed ranking compares recent median generated tokens per second. It favors routes that complete long generations quickly, while first-token latency remains a separate measure of how responsive the endpoint feels at the beginning.
Amazon Bedrock currently ranks second at 71.5 tok/s. Compare that gap with input and output price, response-start time, output speed, 30-minute uptime, context length, and the exact route before deciding whether first place is meaningful for your workload.
Claude Haiku 4.5 has 6 published provider options, and performance data is matched to each exact OpenRouter routing tag. Different quantization, context, regional deployment, or provider configuration can change price and behavior even when the underlying model name is identical.
