DeepSeek V4 Pro 0423
deepseek/deepseek-v4-pro
Compare with DeepSeek V4 Flash 0731, DeepSeek V4 Flash 0423, DeepSeek V3.2
| Provider | State | Uptime 5m | 24h | p50 latency | p99 | Throughput | Context | $/Mtok |
|---|---|---|---|---|---|---|---|---|
BaseTenbaseten/fp4fp4 | Healthy | 100% | 99.13% | 421ms | 4.10s | 67 tps | 1.0M | $1.74 / $3.48 |
DeepInfradeepinfra/fp8fp8 | Healthy | 100% | 96.25% | 1.12s | 42.8s | 21 tps | 1.0M | $1.30 / $2.60 |
DeepSeekdeepseek | Healthy | 100% | 77.77% | 1.24s | 3.86s | 28 tps | 1.0M | $0.660 / $1.98 |
Novitanovita/fp8fp8 | Healthy | 99.97% | 99.10% | 1.63s | 42.3s | 56 tps | 1.0M | $1.44 / $2.88 |
Alibabaalibaba/fp8fp8 | Healthy | 99.91% | 97.80% | 1.56s | 6.69s | 52 tps | 1.0M | $1.42 / $2.83 |
Baidubaidu/fp8fp8 | Healthy | 99.86% | 97.72% | 806ms | 6.80s | 64 tps | 1.0M | $1.69 / $3.38 |
AtlasCloudatlas-cloud/fp4fp4 | Degraded | 99.16% | 92.79% | 1.39s | 6.71s | 38 tps | 1.0M | $1.68 / $3.38 |
Parasailparasail/fp8fp8 | Degraded | 99.14% | 88.75% | 777ms | 8.26s | 50 tps | 1.0M | $1.74 / $3.48 |
GMICloudgmicloud/fp8fp8 | Degraded | 99.14% | 95.69% | 3.26s | 11.7s | 27 tps | 1.0M | $0.696 / $1.39 |
SiliconFlowsiliconflow/fp8fp8 | Degraded | 98.75% | 96.04% | 1.70s | 83.5s | 40 tps | 1.0M | $1.50 / $3.14 |
StreamLakestreamlake/fp8fp8 | Healthy | 97.92% | 96.13% | 2.67s | 22.3s | 36 tps | 1.0M | $0.694 / $1.39 |
Ionstreamionstream/fp4fp4 | Degraded | 97.65% | 95.58% | 3.35s | 15.5s | 15 tps | 1.0M | $1.13 / $2.26 |
Togethertogether | Degraded | 94.66% | 89.84% | 674ms | 14.5s | 51 tps | 512K | $1.74 / $3.48 |
Azureazure/us | Degraded | 94.01% | 93.75% | 1.47s | 14.8s | 36 tps | 1.0M | $1.91 / $3.83 |
CoreWeavecoreweave/fp8fp8 | No data | — | 97.64% | 686ms | 7.91s | 60 tps | 1.0M | $1.15 / $2.55 |
DigitalOceandigitalocean | Down | — | 94.01% | — | — | — | 1.0M | $0.870 / $1.74 |
Fireworksfireworks | No data | — | 3.40% | 2.58s | 17.4s | 43 tps | 1.0M | $1.74 / $3.48 |
Venicevenice | Down | — | 74.17% | 2.92s | 41.9s | 41 tps | 1.0M | $1.65 / $3.30 |
- BaseTen100%Healthy24h 99.13%
- p50
- 421ms
- p99
- 4.10s
- Throughput
- 67 tps
- Context
- 1.0M
- $/Mtok
- $1.74 / $3.48
baseten/fp4fp4 - DeepInfra100%Healthy24h 96.25%
- p50
- 1.12s
- p99
- 42.8s
- Throughput
- 21 tps
- Context
- 1.0M
- $/Mtok
- $1.30 / $2.60
deepinfra/fp8fp8 - DeepSeek100%Healthy24h 77.77%
- p50
- 1.24s
- p99
- 3.86s
- Throughput
- 28 tps
- Context
- 1.0M
- $/Mtok
- $0.660 / $1.98
deepseek - Novita99.97%Healthy24h 99.10%
- p50
- 1.63s
- p99
- 42.3s
- Throughput
- 56 tps
- Context
- 1.0M
- $/Mtok
- $1.44 / $2.88
novita/fp8fp8 - Alibaba99.91%Healthy24h 97.80%
- p50
- 1.56s
- p99
- 6.69s
- Throughput
- 52 tps
- Context
- 1.0M
- $/Mtok
- $1.42 / $2.83
alibaba/fp8fp8 - Baidu99.86%Healthy24h 97.72%
- p50
- 806ms
- p99
- 6.80s
- Throughput
- 64 tps
- Context
- 1.0M
- $/Mtok
- $1.69 / $3.38
baidu/fp8fp8 - AtlasCloud99.16%Degraded24h 92.79%
- p50
- 1.39s
- p99
- 6.71s
- Throughput
- 38 tps
- Context
- 1.0M
- $/Mtok
- $1.68 / $3.38
atlas-cloud/fp4fp4 - Parasail99.14%Degraded24h 88.75%
- p50
- 777ms
- p99
- 8.26s
- Throughput
- 50 tps
- Context
- 1.0M
- $/Mtok
- $1.74 / $3.48
parasail/fp8fp8 - GMICloud99.14%Degraded24h 95.69%
- p50
- 3.26s
- p99
- 11.7s
- Throughput
- 27 tps
- Context
- 1.0M
- $/Mtok
- $0.696 / $1.39
gmicloud/fp8fp8 - SiliconFlow98.75%Degraded24h 96.04%
- p50
- 1.70s
- p99
- 83.5s
- Throughput
- 40 tps
- Context
- 1.0M
- $/Mtok
- $1.50 / $3.14
siliconflow/fp8fp8 - StreamLake97.92%Healthy24h 96.13%
- p50
- 2.67s
- p99
- 22.3s
- Throughput
- 36 tps
- Context
- 1.0M
- $/Mtok
- $0.694 / $1.39
streamlake/fp8fp8 - Ionstream97.65%Degraded24h 95.58%
- p50
- 3.35s
- p99
- 15.5s
- Throughput
- 15 tps
- Context
- 1.0M
- $/Mtok
- $1.13 / $2.26
ionstream/fp4fp4 - Together94.66%Degraded24h 89.84%
- p50
- 674ms
- p99
- 14.5s
- Throughput
- 51 tps
- Context
- 512K
- $/Mtok
- $1.74 / $3.48
together - Azure94.01%Degraded24h 93.75%
- p50
- 1.47s
- p99
- 14.8s
- Throughput
- 36 tps
- Context
- 1.0M
- $/Mtok
- $1.91 / $3.83
azure/us - CoreWeave—No data24h 97.64%
- p50
- 686ms
- p99
- 7.91s
- Throughput
- 60 tps
- Context
- 1.0M
- $/Mtok
- $1.15 / $2.55
coreweave/fp8fp8 - DigitalOcean—Down24h 94.01%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $0.870 / $1.74
digitalocean - Fireworks—No data24h 3.40%
- p50
- 2.58s
- p99
- 17.4s
- Throughput
- 43 tps
- Context
- 1.0M
- $/Mtok
- $1.74 / $3.48
fireworks - Venice—Down24h 74.17%
- p50
- 2.92s
- p99
- 41.9s
- Throughput
- 41 tps
- Context
- 1.0M
- $/Mtok
- $1.65 / $3.30
venice
Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
History
2 endpoints reporting no data are omitted from the charts.
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 and p99 are OpenRouter's rolling 30-minute windows.
Throughput
Median tokens per second.