R1 0528: providers compared
Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.
Current state
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100%◆ (best) | 99.95% | 99.94%◆ (best) | 99.94%◆ (best) | 794ms◆ (best) | 3.40s◆ (best) | 20 tps | 164K | $0.500 / $2.15◆ (best) | |
| Healthy | 100%◆ (best) | 99.90% | 99.85% | 99.83% | 1.19s | 6.11s | 21 tps | 164K | $0.700 / $2.50 | |
| Healthy | 100%◆ (best) | 99.07% | 99.54% | 99.40% | 1.90s | 7.84s | 20 tps | 164K | $0.500 / $2.18◆ (best) | |
StreamLakestreamlake | Healthy | 100%◆ (best) | 99.35% | 99.27% | 99.41% | 3.35s | 9.51s | 48 tps◆ (best) | 128K | $0.571 / $2.29 |
- DeepInfra100%◆ (best)Healthy24h 99.95%
- 7d
- 99.94%◆ (best)
- 30d
- 99.94%◆ (best)
- p50
- 794ms◆ (best)
- p99
- 3.40s◆ (best)
- Throughput
- 20 tps
- Context
- 164K
- $/Mtok
- $0.500 / $2.15◆ (best)
deepinfra/fp4fp4 - Novita100%◆ (best)Healthy24h 99.90%
- 7d
- 99.85%
- 30d
- 99.83%
- p50
- 1.19s
- p99
- 6.11s
- Throughput
- 21 tps
- Context
- 164K
- $/Mtok
- $0.700 / $2.50
novita/fp8fp8 - SiliconFlow100%◆ (best)Healthy24h 99.07%
- 7d
- 99.54%
- 30d
- 99.40%
- p50
- 1.90s
- p99
- 7.84s
- Throughput
- 20 tps
- Context
- 164K
- $/Mtok
- $0.500 / $2.18◆ (best)
siliconflow/fp8fp8 - StreamLake100%◆ (best)Healthy24h 99.35%
- 7d
- 99.27%
- 30d
- 99.41%
- p50
- 3.35s
- p99
- 9.51s
- Throughput
- 48 tps◆ (best)
- Context
- 128K
- $/Mtok
- $0.571 / $2.29
streamlake
Measured 1m 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.