Hy4 preview: 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) | 80.72% | 95.08% | 96.92% | 4.03s | 15.5s | 51 tps | 1.0M | $0.834 / $2.50 | |
| Healthy | 100%◆ (best) | 95.49% | 99.11% | 98.95% | 1.51s | 2.90s◆ (best) | 65 tps◆ (best) | 1.0M | $0.834 / $2.50 | |
| Healthy | 99.29% | 98.01% | 96.98% | 98.98% | 3.92s | 11.6s | 50 tps | 1.0M | $0.751 / $2.25 | |
| No data | — | 98.19% | 99.75%◆ (best) | 99.75%◆ (best) | 1.45s◆ (best) | 5.33s | 23 tps | 1.0M | $0.834 / $2.50 | |
| No data | — | — | 95.81% | 95.81% | — | — | — | 1.0M | $0.747 / $2.25◆ (best) |
- Novita100%◆ (best)Healthy24h 80.72%
- 7d
- 95.08%
- 30d
- 96.92%
- p50
- 4.03s
- p99
- 15.5s
- Throughput
- 51 tps
- Context
- 1.0M
- $/Mtok
- $0.834 / $2.50
novita/fp8fp8 - SiliconFlow100%◆ (best)Healthy24h 95.49%
- 7d
- 99.11%
- 30d
- 98.95%
- p50
- 1.51s
- p99
- 2.90s◆ (best)
- Throughput
- 65 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $0.834 / $2.50
siliconflow/fp8fp8 - Tencent99.29%Healthy24h 98.01%
- 7d
- 96.98%
- 30d
- 98.98%
- p50
- 3.92s
- p99
- 11.6s
- Throughput
- 50 tps
- Context
- 1.0M
- $/Mtok
- $0.751 / $2.25
tencent/fp8fp8 - No data24h 98.19%
- 7d
- 99.75%◆ (best)
- 30d
- 99.75%◆ (best)
- p50
- 1.45s◆ (best)
- p99
- 5.33s
- Throughput
- 23 tps
- Context
- 1.0M
- $/Mtok
- $0.834 / $2.50
deepinfra/fp8fp8 - No data24h —
- 7d
- 95.81%
- 30d
- 95.81%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $0.747 / $2.25◆ (best)
inference-net/fp8fp8
Measured 3m 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.
Throughput
Median tokens per second.