Start typing to search.
Ranks providers by estimated time to return 500 tokens, combining the wait for the first token with output speed.
Groq currently ranks first at 3.76 s via groq.
| Rank | Provider / route | Estimated response time | Speed measurements | Endpoint details |
|---|---|---|---|---|
| #1 | Provider / route Groq groq | Estimated response time3.76 s | Speed measurements
| Endpoint details
|
| #2 | Provider / route Weights & Biases wandb/fp16 | Estimated response time9.52 s | Speed measurements
| Endpoint details
|
| #3 | Provider / route SambaNova sambanova-turbo | Estimated response time10.06 s | Speed measurements
| Endpoint details
|
| #4 | Provider / route Google Vertex google-vertex | Estimated response time10.53 s | Speed measurements
| Endpoint details
|
| #5 | Provider / route Parasail parasail/fp8 | Estimated response time11.59 s | Speed measurements
| Endpoint details
|
| #6 | Provider / route Cloudflare cloudflare/fp8 | Estimated response time17.72 s | Speed measurements
| Endpoint details
|
| #7 | Provider / route AkashML akashml/fp8 | Estimated response time18.97 s | Speed measurements
| Endpoint details
|
| #8 | Provider / route Nebius nebius/fp8 | Estimated response time20.36 s | Speed measurements
| Endpoint details
|
| #9 | Provider / route Together together/fp8 | Estimated response time21.65 s | Speed measurements
| Endpoint details
|
| #10 | Provider / route Google Vertex google-vertex/us-central1 | Estimated response time31.67 s | Speed measurements
| Endpoint details
|
| #11 | Provider / route DeepInfra deepinfra/turbo | Estimated response time71.84 s | Speed measurements
| Endpoint details
|
| #12 | Provider / route NovitaAI novita/bf16 | Estimated response time100.67 s | Speed measurements
| Endpoint details
|
Endpoint data fetched .
Llama 3.3 70B Instruct endpoint guide
12 of 12 Llama 3.3 70B Instruct endpoints currently have the published data required for this ranking. Groq leads at 3.76 s via groq. The table keeps unranked routes visible so missing measurements do not look like missing provider availability.
The fastest ranking estimates a 500-token answer by adding the recent median first-token wait to 500 divided by median output throughput. It represents an example complete response, not a guarantee for every prompt or request size.
Weights & Biases currently ranks second at 9.52 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.
Llama 3.3 70B Instruct has 12 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.
