DeepSeek V4 Pro 0813: 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Alibabaalibaba | Healthy | 100%◆ (best) | 99.83% | 99.62% | 99.62% | 1.51s | 16.3s | 52 tps | 1.0M | $0.603 / $1.81◆ (best) |
DeepSeekdeepseek | Healthy | 100%◆ (best) | 99.95% | 86.81% | 86.81% | 2.10s | 25.5s | 32 tps | 1.0M | $0.660 / $1.98 |
Fireworksfireworks | Healthy | 100%◆ (best) | 98.44% | 99.23% | 99.23% | 1.30s | 8.74s◆ (best) | 44 tps | 1.0M | $1.32 / $3.96 |
| Healthy | 100%◆ (best) | 99.59% | 99.67%◆ (best) | 99.67%◆ (best) | 4.76s | 31.2s | 42 tps | 1.0M | $1.19 / $3.56 | |
| Healthy | 100%◆ (best) | 99.47% | 99.03% | 99.03% | 2.22s | 25.4s | 54 tps | 1.0M | $1.32 / $3.96 | |
| Healthy | 100%◆ (best) | 96.70% | 95.97% | 95.97% | 730ms◆ (best) | 15.0s | 69 tps | 1.0M | $1.32 / $3.96 | |
| Healthy | 100%◆ (best) | 99.72% | 99.50% | 99.50% | 785ms | 10.6s | 115 tps◆ (best) | 1.0M | $1.32 / $3.96 | |
| Degraded | 100%◆ (best) | 98.41% | 99.12% | 99.12% | 3.05s | 134s | 41 tps | 1.0M | $1.32 / $3.96 | |
StreamLakestreamlake | Healthy | 100%◆ (best) | 98.25% | 98.72% | 98.72% | 3.01s | 80.5s | 30 tps | 1.0M | $1.32 / $3.96 |
Togethertogether | Healthy | 100%◆ (best) | 97.44% | 97.17% | 97.17% | 1.14s | 16.0s | 73 tps | 1.0M | $1.32 / $3.96 |
Cloudflarecloudflare | Degraded | 97.96% | 97.97% | 97.21% | 97.21% | 4.20s | 69.9s | 37 tps | 1.0M | $1.32 / $3.96 |
| No data | — | 99.00% | 98.67% | 98.67% | 1.04s | 15.7s | 102 tps | 1.0M | $1.32 / $3.96 | |
DigitalOceandigitalocean | No data | — | 97.82% | 97.04% | 97.04% | 1.23s | 17.0s | 47 tps | 1.0M | $1.32 / $3.96 |
- Alibaba100%◆ (best)Healthy24h 99.83%
- 7d
- 99.62%
- 30d
- 99.62%
- p50
- 1.51s
- p99
- 16.3s
- Throughput
- 52 tps
- Context
- 1.0M
- $/Mtok
- $0.603 / $1.81◆ (best)
alibaba - DeepSeek100%◆ (best)Healthy24h 99.95%
- 7d
- 86.81%
- 30d
- 86.81%
- p50
- 2.10s
- p99
- 25.5s
- Throughput
- 32 tps
- Context
- 1.0M
- $/Mtok
- $0.660 / $1.98
deepseek - Fireworks100%◆ (best)Healthy24h 98.44%
- 7d
- 99.23%
- 30d
- 99.23%
- p50
- 1.30s
- p99
- 8.74s◆ (best)
- Throughput
- 44 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
fireworks - GMICloud100%◆ (best)Healthy24h 99.59%
- 7d
- 99.67%◆ (best)
- 30d
- 99.67%◆ (best)
- p50
- 4.76s
- p99
- 31.2s
- Throughput
- 42 tps
- Context
- 1.0M
- $/Mtok
- $1.19 / $3.56
gmicloud/fp8fp8 - Novita100%◆ (best)Healthy24h 99.47%
- 7d
- 99.03%
- 30d
- 99.03%
- p50
- 2.22s
- p99
- 25.4s
- Throughput
- 54 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
novita/fp8fp8 - Parasail100%◆ (best)Healthy24h 96.70%
- 7d
- 95.97%
- 30d
- 95.97%
- p50
- 730ms◆ (best)
- p99
- 15.0s
- Throughput
- 69 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
parasail/fp8fp8 - Sail Research100%◆ (best)Healthy24h 99.72%
- 7d
- 99.50%
- 30d
- 99.50%
- p50
- 785ms
- p99
- 10.6s
- Throughput
- 115 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
sail-research/fp4fp4 - SiliconFlow100%◆ (best)Degraded24h 98.41%
- 7d
- 99.12%
- 30d
- 99.12%
- p50
- 3.05s
- p99
- 134s
- Throughput
- 41 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
siliconflow/fp8fp8 - StreamLake100%◆ (best)Healthy24h 98.25%
- 7d
- 98.72%
- 30d
- 98.72%
- p50
- 3.01s
- p99
- 80.5s
- Throughput
- 30 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
streamlake - Together100%◆ (best)Healthy24h 97.44%
- 7d
- 97.17%
- 30d
- 97.17%
- p50
- 1.14s
- p99
- 16.0s
- Throughput
- 73 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
together - Cloudflare97.96%Degraded24h 97.97%
- 7d
- 97.21%
- 30d
- 97.21%
- p50
- 4.20s
- p99
- 69.9s
- Throughput
- 37 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
cloudflare - No data24h 99.00%
- 7d
- 98.67%
- 30d
- 98.67%
- p50
- 1.04s
- p99
- 15.7s
- Throughput
- 102 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
baseten/fp4fp4 - No data24h 97.82%
- 7d
- 97.04%
- 30d
- 97.04%
- p50
- 1.23s
- p99
- 17.0s
- Throughput
- 47 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
digitalocean
Measured 5m 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.