Start typing to search.
Ranks providers by how quickly the first generated token appears. Providers without enough recent data remain visible after ranked rows.
Baseten currently ranks first at 0.50 s via baseten/fp4.
| Rank | Provider / route | Time to first token | Speed measurements | Endpoint details |
|---|---|---|---|---|
| #1 | Provider / route Baseten baseten/fp4 | Time to first token0.50 s | Speed measurements
| Endpoint details
|
| #2 | Provider / route Baidu Qianfan baidu/fp8 | Time to first token0.74 s | Speed measurements
| Endpoint details
|
| #3 | Provider / route DeepInfra deepinfra/fp4 | Time to first token0.79 s | Speed measurements
| Endpoint details
|
| #4 | Provider / route Together together | Time to first token0.79 s | Speed measurements
| Endpoint details
|
| #5 | Provider / route DigitalOcean digitalocean | Time to first token0.80 s | Speed measurements
| Endpoint details
|
| #6 | Provider / route Weights & Biases wandb/fp8 | Time to first token1.18 s | Speed measurements
| Endpoint details
|
| #7 | Provider / route SiliconFlow siliconflow/fp8 | Time to first token1.21 s | Speed measurements
| Endpoint details
|
| #8 | Provider / route DeepSeek deepseek | Time to first token1.27 s | Speed measurements
| Endpoint details
|
| #9 | Provider / route Alibaba Cloud Int. alibaba | Time to first token1.38 s | Speed measurements
| Endpoint details
|
| #10 | Provider / route Fireworks fireworks | Time to first token1.43 s | Speed measurements
| Endpoint details
|
| #11 | Provider / route NovitaAI novita/fp8 | Time to first token1.47 s | Speed measurements
| Endpoint details
|
| #12 | Provider / route AtlasCloud atlas-cloud/fp4 | Time to first token2.06 s | Speed measurements
| Endpoint details
|
| #13 | Provider / route Venice venice | Time to first token2.24 s | Speed measurements
| Endpoint details
|
| #14 | Provider / route StreamLake streamlake/fp8 | Time to first token2.43 s | Speed measurements
| Endpoint details
|
| #15 | Provider / route GMICloud gmicloud/fp8 | Time to first token2.50 s | Speed measurements
| Endpoint details
|
| — | Provider / route Parasail parasail/fp8 | Time to first tokenInsufficient recent data | Speed measurements
| Endpoint details
|
Endpoint data fetched .
DeepSeek V4 Pro endpoint guide
15 of 16 DeepSeek V4 Pro endpoints currently have the published data required for this ranking. Baseten leads at 0.50 s via baseten/fp4. The table keeps unranked routes visible so missing measurements do not look like missing provider availability.
The first-token ranking orders endpoints by the recent median time before output begins. This is useful for interactive applications, but it should be read alongside throughput because the first provider to start is not always the first to finish a long answer.
Baidu Qianfan currently ranks second at 0.74 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.
DeepSeek V4 Pro has 16 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.
