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.
DeepInfra currently ranks first at 53.0 tok/s via deepinfra/fp8.
| Rank | Provider / route | Output speed | Speed measurements | Endpoint details |
|---|---|---|---|---|
| #1 | Provider / route DeepInfra deepinfra/fp8 | Output speed53.0 tok/s | Speed measurements
| Endpoint details
|
| #2 | Provider / route GMICloud gmicloud/bf16 | Output speed46.0 tok/s | Speed measurements
| Endpoint details
|
| #3 | Provider / route Xiaomi xiaomi/fp8 | Output speed44.0 tok/s | Speed measurements
| Endpoint details
|
| #4 | Provider / route NovitaAI novita | Output speed39.0 tok/s | Speed measurements
| Endpoint details
|
| #5 | Provider / route DigitalOcean digitalocean | Output speed36.0 tok/s | Speed measurements
| Endpoint details
|
| #6 | Provider / route AtlasCloud atlas-cloud/fp8 | Output speed25.0 tok/s | Speed measurements
| Endpoint details
|
Endpoint data fetched .
MiMo-V2.5-Pro endpoint guide
6 of 6 MiMo-V2.5-Pro endpoints currently have the published data required for this ranking. DeepInfra leads at 53.0 tok/s via deepinfra/fp8. 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.
GMICloud currently ranks second at 46.0 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.
MiMo-V2.5-Pro 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.
