DeepSeek V4 Pro 0813
deepseek/deepseek-v4-pro-0813
Compare with DeepSeek V4 Pro 0423, DeepSeek V4 Flash 0731, DeepSeek V4 Flash 0423
CompareDeepSeek V4 Pro 0813
| State | ||||||||
|---|---|---|---|---|---|---|---|---|
| Healthy | 100% | 97.14% | 1.01s | 21.1s | 108 tps | 1.0M | $1.32 / $3.96 | |
Cloudflarecloudflare | Healthy | 100% | 98.14% | 2.38s | 48.4s | 42 tps | 1.0M | $1.32 / $3.96 |
DeepSeekdeepseek | Healthy | 100% | 99.97% | 1.44s | 8.50s | 29 tps | 1.0M | $0.660 / $1.98 |
DigitalOceandigitalocean | Degraded | 100% | 97.20% | 1.40s | 15.3s | 49 tps | 1.0M | $1.32 / $3.96 |
Fireworksfireworks | Healthy | 100% | 98.39% | 2.74s | 15.9s | 40 tps | 1.0M | $1.32 / $3.96 |
| Healthy | 100% | 99.60% | 1.73s | 34.9s | 49 tps | 1.0M | $1.32 / $3.96 | |
StreamLakestreamlake | Healthy | 100% | 99.19% | 1.80s | 16.4s | 48 tps | 1.0M | $1.32 / $3.96 |
Togethertogether | Healthy | 100% | 97.60% | 1.73s | 64.2s | 56 tps | 1.0M | $1.32 / $3.96 |
Alibabaalibaba | Healthy | 99.97% | 99.85% | 1.27s | 6.46s | 59 tps | 1.0M | $0.603 / $1.81 |
| Healthy | 99.80% | 99.74% | 4.57s | 33.3s | 39 tps | 1.0M | $1.19 / $3.56 | |
| Healthy | 98.53% | 99.83% | 1.57s | 15.4s | 50 tps | 1.0M | $1.32 / $3.96 | |
| Degraded | 93.55% | 95.83% | 981ms | 14.2s | 67 tps | 1.0M | $1.32 / $3.96 |
- BaseTen100%Healthy24h 97.14%
- p50
- 1.01s
- p99
- 21.1s
- Throughput
- 108 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
baseten/fp4fp4 - Cloudflare100%Healthy24h 98.14%
- p50
- 2.38s
- p99
- 48.4s
- Throughput
- 42 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
cloudflare - DeepSeek100%Healthy24h 99.97%
- p50
- 1.44s
- p99
- 8.50s
- Throughput
- 29 tps
- Context
- 1.0M
- $/Mtok
- $0.660 / $1.98
deepseek - DigitalOcean100%Degraded24h 97.20%
- p50
- 1.40s
- p99
- 15.3s
- Throughput
- 49 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
digitalocean - Fireworks100%Healthy24h 98.39%
- p50
- 2.74s
- p99
- 15.9s
- Throughput
- 40 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
fireworks - SiliconFlow100%Healthy24h 99.60%
- p50
- 1.73s
- p99
- 34.9s
- Throughput
- 49 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
siliconflow/fp8fp8 - StreamLake100%Healthy24h 99.19%
- p50
- 1.80s
- p99
- 16.4s
- Throughput
- 48 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
streamlake - Together100%Healthy24h 97.60%
- p50
- 1.73s
- p99
- 64.2s
- Throughput
- 56 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
together - Alibaba99.97%Healthy24h 99.85%
- p50
- 1.27s
- p99
- 6.46s
- Throughput
- 59 tps
- Context
- 1.0M
- $/Mtok
- $0.603 / $1.81
alibaba - GMICloud99.80%Healthy24h 99.74%
- p50
- 4.57s
- p99
- 33.3s
- Throughput
- 39 tps
- Context
- 1.0M
- $/Mtok
- $1.19 / $3.56
gmicloud/fp8fp8 - Novita98.53%Healthy24h 99.83%
- p50
- 1.57s
- p99
- 15.4s
- Throughput
- 50 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
novita/fp8fp8 - Parasail93.55%Degraded24h 95.83%
- p50
- 981ms
- p99
- 14.2s
- Throughput
- 67 tps
- Context
- 1.0M
- $/Mtok
- $1.32 / $3.96
parasail/fp8fp8
Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
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.