Llama 4 Maverick: 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
DigitalOceandigitalocean | Healthy | 100%◆ (best) | 99.28% | 99.49% | 99.71% | 733ms | 9.65s | 11 tps | 128K | $0.188 / $0.652◆ (best) |
| Healthy | 100%◆ (best) | 99.84% | 99.83%◆ (best) | 99.80%◆ (best) | 527ms◆ (best) | 6.44s◆ (best) | 53 tps◆ (best) | 524K | $0.350 / $1.00 | |
| Healthy | 98.77% | 98.73% | 98.27% | 99.29% | 721ms | 7.16s | 17 tps | 1.0M | $0.270 / $0.850 | |
| No data | — | — | 99.76% | 99.66% | — | — | — | 1.0M | $0.200 / $0.800 | |
Googlegoogle-vertex/us-east5 | No data | — | — | — | 99.39% | — | — | — | 524K | $0.350 / $1.15 |
- DigitalOcean100%◆ (best)Healthy24h 99.28%
- 7d
- 99.49%
- 30d
- 99.71%
- p50
- 733ms
- p99
- 9.65s
- Throughput
- 11 tps
- Context
- 128K
- $/Mtok
- $0.188 / $0.652◆ (best)
digitalocean - Parasail100%◆ (best)Healthy24h 99.84%
- 7d
- 99.83%◆ (best)
- 30d
- 99.80%◆ (best)
- p50
- 527ms◆ (best)
- p99
- 6.44s◆ (best)
- Throughput
- 53 tps◆ (best)
- Context
- 524K
- $/Mtok
- $0.350 / $1.00
parasail/fp8fp8 - Novita98.77%Healthy24h 98.73%
- 7d
- 98.27%
- 30d
- 99.29%
- p50
- 721ms
- p99
- 7.16s
- Throughput
- 17 tps
- Context
- 1.0M
- $/Mtok
- $0.270 / $0.850
novita/fp8fp8 - No data24h —
- 7d
- 99.76%
- 30d
- 99.66%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $0.200 / $0.800
deepinfra/basefp8 - No data24h —
- 7d
- —
- 30d
- 99.39%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 524K
- $/Mtok
- $0.350 / $1.15
google-vertex/us-east5
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.
Throughput
Median tokens per second.