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.
Venice currently ranks first at 113.5 tok/s via venice.
| Rank | Provider / route | Output speed | Speed measurements | Endpoint details |
|---|---|---|---|---|
| #1 | Provider / route Venice venice | Output speed113.5 tok/s | Speed measurements
| Endpoint details
|
| #2 | Provider / route DeepInfra deepinfra/fp4 | Output speed67.0 tok/s | Speed measurements
| Endpoint details
|
| #3 | Provider / route StreamLake streamlake/fp8 | Output speed50.0 tok/s | Speed measurements
| Endpoint details
|
| #4 | Provider / route Phala phala | Output speed49.0 tok/s | Speed measurements
| Endpoint details
|
| #5 | Provider / route Moonshot AI moonshotai/int4 | Output speed41.0 tok/s | Speed measurements
| Endpoint details
|
| #6 | Provider / route SiliconFlow siliconflow/int4 | Output speed40.0 tok/s | Speed measurements
| Endpoint details
|
| #7 | Provider / route NovitaAI novita | Output speed33.0 tok/s | Speed measurements
| Endpoint details
|
| #8 | Provider / route ModelRun modelrun/fp4 | Output speed30.0 tok/s | Speed measurements
| Endpoint details
|
| #9 | Provider / route AtlasCloud atlas-cloud/int4 | Output speed29.0 tok/s | Speed measurements
| Endpoint details
|
| #10 | Provider / route Chutes chutes/int4 | Output speed17.0 tok/s | Speed measurements
| Endpoint details
|
| #11 | Provider / route DigitalOcean digitalocean | Output speed12.0 tok/s | Speed measurements
| Endpoint details
|
Endpoint data fetched .
Kimi K2.5 endpoint guide
11 of 11 Kimi K2.5 endpoints currently have the published data required for this ranking. Venice leads at 113.5 tok/s via venice. 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.
DeepInfra currently ranks second at 67.0 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.
Kimi K2.5 has 11 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.
