Qwen2.5 72B Instruct: providers compared
Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.
Current state
- DeepInfra100%◆ (best)Healthy24h 99.51%
- 7d
- 97.32%◆ (best)
- 30d
- 98.99%◆ (best)
- p50
- 322ms◆ (best)
- p99
- 4.24s
- Throughput
- 27 tps◆ (best)
- Context
- 33K
- $/Mtok
- $0.360 / $0.400◆ (best)
deepinfra/fp8fp8 - Novita100%◆ (best)Healthy24h 98.44%
- 7d
- 95.58%
- 30d
- 95.25%
- p50
- 519ms
- p99
- 3.02s◆ (best)
- Throughput
- 26 tps
- Context
- 32K
- $/Mtok
- $0.380 / $0.400
novita/bf16bf16
Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.
Price vs latency
One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag
History
Availability
Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.
Time to first token
Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window, averaged over the hour.
Throughput
Median tokens per second.