Step 3.7 Flash: 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% | 99.75% | 99.81% | 99.37% | 4.99s | 19.2s | 36 tps | 262K | $0.200 / $1.15 | |
DeepInfradeepinfra | No data | — | — | — | 99.93% | — | — | — | 262K | $0.160 / $0.920◆ (best) |
DeepInfradeepinfra/us | No data | — | — | 99.98%◆ (best) | 99.95%◆ (best) | — | — | — | 262K | $0.160 / $0.920◆ (best) |
| No data | — | 99.64% | 99.74% | 99.37% | 4.08s◆ (best) | 19.1s◆ (best) | 73 tps◆ (best) | 256K | $0.200 / $1.15 |
- Novita100%Healthy24h 99.75%
- 7d
- 99.81%
- 30d
- 99.37%
- p50
- 4.99s
- p99
- 19.2s
- Throughput
- 36 tps
- Context
- 262K
- $/Mtok
- $0.200 / $1.15
novita/fp8fp8 - No data24h —
- 7d
- —
- 30d
- 99.93%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.160 / $0.920◆ (best)
deepinfra - No data24h —
- 7d
- 99.98%◆ (best)
- 30d
- 99.95%◆ (best)
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.160 / $0.920◆ (best)
deepinfra/us - No data24h 99.64%
- 7d
- 99.74%
- 30d
- 99.37%
- p50
- 4.08s◆ (best)
- p99
- 19.1s◆ (best)
- Throughput
- 73 tps◆ (best)
- Context
- 256K
- $/Mtok
- $0.200 / $1.15
stepfun/fp8fp8
Measured 4m 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.