DeepSeek V4 Flash 0731
deepseek/deepseek-v4-flash-0731
Compare with DeepSeek V4 Flash 0423, DeepSeek V4 Pro 0423, DeepSeek V4 Pro 0813
CompareDeepSeek V4 Flash 0731
| Provider | State | Uptime 5m | 24h | p50 latency | p99 | Throughput | Context | $/Mtok |
|---|---|---|---|---|---|---|---|---|
BaseTenbaseten/fp8fp8 | Healthy | 100% | 98.38% | 619ms | 7.92s | 50 tps | 1.0M | $0.130 / $0.260 |
DeepSeekdeepseek | Healthy | 100% | 67.55% | 820ms | 2.64s | 72 tps | 1.0M | $0.220 / $0.660 |
Novitanovita/fp8fp8 | Healthy | 100% | 99.33% | 1.22s | 5.35s | 91 tps | 1.0M | $0.140 / $0.280 |
CoreWeavecoreweave/fp8fp8 | Healthy | 99.97% | 99.89% | 650ms | 7.61s | 108 tps | 262K | $0.130 / $0.280 |
GMICloudgmicloud/fp8fp8 | Healthy | 99.96% | 98.61% | 2.43s | 10.0s | 52 tps | 1.0M | $0.112 / $0.224 |
Baidubaidu/fp8fp8 | Healthy | 99.84% | 99.24% | 976ms | 7.66s | 113 tps | 1.0M | $0.111 / $0.221 |
Cloudflarecloudflare | Healthy | 99.83% | 99.94% | 743ms | 7.70s | 51 tps | 1.3M | $0.440 / $1.32 |
Relacerelace/fp4fp4 | Healthy | 99.68% | 99.34% | 499ms | 3.84s | 73 tps | 1.0M | $0.105 / $0.210 |
Inceptroninceptron/fp4fp4 | Healthy | 99.65% | 98.92% | 661ms | 9.53s | 33 tps | 1.0M | $0.130 / $0.280 |
Io Netio-net/fp8fp8 | Healthy | 99.47% | 98.67% | 1.67s | 18.5s | 32 tps | 262K | $0.149 / $0.320 |
Morphmorph/bf16bf16 | Healthy | 99.42% | 97.76% | 1.58s | 32.8s | 15 tps | 1.0M | $0.139 / $0.278 |
Ambientambient/fp4fp4 | Degraded | 99.36% | 92.18% | 1.37s | 16.4s | 37 tps | 1.0M | $0.140 / $0.280 |
AtlasCloudatlas-cloud/fp4fp4 | Healthy | 99.31% | 99.33% | 2.08s | 9.20s | 100 tps | 1.0M | $0.140 / $0.280 |
DeepInfradeepinfra/fp8fp8 | Healthy | 99.30% | 98.06% | 843ms | 23.9s | 24 tps | 1.0M | $0.080 / $0.180 |
Sail Researchsail-research/fp4fp4 | Healthy | 99.16% | 99.13% | 1.09s | 41.3s | 29 tps | 1.0M | $0.090 / $0.180 |
DigitalOceandigitalocean | Healthy | 98.86% | 95.71% | 628ms | 7.86s | 23 tps | 1.0M | $0.080 / $0.252 |
Decartdecart/fp4fp4 | Healthy | 98.76% | 73.98% | 711ms | 8.86s | 37 tps | 262K | $0.076 / $0.153 |
Fireworksfireworks | Healthy | 98.68% | 93.01% | 834ms | 32.2s | 91 tps | 1.0M | $0.140 / $0.280 |
OpenInferenceopen-inference/fp4fp4 | Healthy | 98.45% | 95.75% | 4.97s | 32.1s | 9.0 tps | 262K | $0.080 / $0.180 |
Waferwafer/fast | Healthy | 98.23% | 96.56% | 1.23s | 20.7s | 129 tps | 1.0M | $0.280 / $0.560 |
Phalaphala | Healthy | 98.16% | 96.59% | 1.38s | 16.7s | 31 tps | 1.0M | $0.200 / $0.400 |
Togethertogether | Healthy | 97.74% | 96.47% | 869ms | 13.0s | 48 tps | 1.0M | $0.140 / $0.280 |
Venicevenice | Healthy | 97.03% | 96.03% | 879ms | 14.4s | 64 tps | 1.0M | $0.175 / $0.350 |
StreamLakestreamlake/fp8fp8 | Degraded | 95.71% | 95.24% | 2.58s | 36.3s | 40 tps | 1.0M | $0.079 / $0.157 |
SiliconFlowsiliconflow/fp8fp8 | Healthy | 95.65% | 98.01% | 1.93s | 28.9s | 64 tps | 1.0M | $0.140 / $0.280 |
Parasailparasail/fp8fp8 | Healthy | 95.53% | 98.47% | 756ms | 13.5s | 52 tps | 1.0M | $0.140 / $0.280 |
Mancer 2mancer/fp8fp8 | Degraded | 93.02% | 95.11% | 1.09s | 15.9s | 31 tps | 1.0M | $0.145 / $0.450 |
AkashMLakashml/fp8fp8 | Down | 85.39% | 97.94% | 807ms | 9.66s | 30 tps | 131K | $0.140 / $0.280 |
- BaseTen100%Healthy24h 98.38%
- p50
- 619ms
- p99
- 7.92s
- Throughput
- 50 tps
- Context
- 1.0M
- $/Mtok
- $0.130 / $0.260
baseten/fp8fp8 - DeepSeek100%Healthy24h 67.55%
- p50
- 820ms
- p99
- 2.64s
- Throughput
- 72 tps
- Context
- 1.0M
- $/Mtok
- $0.220 / $0.660
deepseek - Novita100%Healthy24h 99.33%
- p50
- 1.22s
- p99
- 5.35s
- Throughput
- 91 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
novita/fp8fp8 - CoreWeave99.97%Healthy24h 99.89%
- p50
- 650ms
- p99
- 7.61s
- Throughput
- 108 tps
- Context
- 262K
- $/Mtok
- $0.130 / $0.280
coreweave/fp8fp8 - GMICloud99.96%Healthy24h 98.61%
- p50
- 2.43s
- p99
- 10.0s
- Throughput
- 52 tps
- Context
- 1.0M
- $/Mtok
- $0.112 / $0.224
gmicloud/fp8fp8 - Baidu99.84%Healthy24h 99.24%
- p50
- 976ms
- p99
- 7.66s
- Throughput
- 113 tps
- Context
- 1.0M
- $/Mtok
- $0.111 / $0.221
baidu/fp8fp8 - Cloudflare99.83%Healthy24h 99.94%
- p50
- 743ms
- p99
- 7.70s
- Throughput
- 51 tps
- Context
- 1.3M
- $/Mtok
- $0.440 / $1.32
cloudflare - Relace99.68%Healthy24h 99.34%
- p50
- 499ms
- p99
- 3.84s
- Throughput
- 73 tps
- Context
- 1.0M
- $/Mtok
- $0.105 / $0.210
relace/fp4fp4 - Inceptron99.65%Healthy24h 98.92%
- p50
- 661ms
- p99
- 9.53s
- Throughput
- 33 tps
- Context
- 1.0M
- $/Mtok
- $0.130 / $0.280
inceptron/fp4fp4 - Io Net99.47%Healthy24h 98.67%
- p50
- 1.67s
- p99
- 18.5s
- Throughput
- 32 tps
- Context
- 262K
- $/Mtok
- $0.149 / $0.320
io-net/fp8fp8 - Morph99.42%Healthy24h 97.76%
- p50
- 1.58s
- p99
- 32.8s
- Throughput
- 15 tps
- Context
- 1.0M
- $/Mtok
- $0.139 / $0.278
morph/bf16bf16 - Ambient99.36%Degraded24h 92.18%
- p50
- 1.37s
- p99
- 16.4s
- Throughput
- 37 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
ambient/fp4fp4 - AtlasCloud99.31%Healthy24h 99.33%
- p50
- 2.08s
- p99
- 9.20s
- Throughput
- 100 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
atlas-cloud/fp4fp4 - DeepInfra99.30%Healthy24h 98.06%
- p50
- 843ms
- p99
- 23.9s
- Throughput
- 24 tps
- Context
- 1.0M
- $/Mtok
- $0.080 / $0.180
deepinfra/fp8fp8 - Sail Research99.16%Healthy24h 99.13%
- p50
- 1.09s
- p99
- 41.3s
- Throughput
- 29 tps
- Context
- 1.0M
- $/Mtok
- $0.090 / $0.180
sail-research/fp4fp4 - DigitalOcean98.86%Healthy24h 95.71%
- p50
- 628ms
- p99
- 7.86s
- Throughput
- 23 tps
- Context
- 1.0M
- $/Mtok
- $0.080 / $0.252
digitalocean - Decart98.76%Healthy24h 73.98%
- p50
- 711ms
- p99
- 8.86s
- Throughput
- 37 tps
- Context
- 262K
- $/Mtok
- $0.076 / $0.153
decart/fp4fp4 - Fireworks98.68%Healthy24h 93.01%
- p50
- 834ms
- p99
- 32.2s
- Throughput
- 91 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
fireworks - OpenInference98.45%Healthy24h 95.75%
- p50
- 4.97s
- p99
- 32.1s
- Throughput
- 9.0 tps
- Context
- 262K
- $/Mtok
- $0.080 / $0.180
open-inference/fp4fp4 - Wafer98.23%Healthy24h 96.56%
- p50
- 1.23s
- p99
- 20.7s
- Throughput
- 129 tps
- Context
- 1.0M
- $/Mtok
- $0.280 / $0.560
wafer/fast - Phala98.16%Healthy24h 96.59%
- p50
- 1.38s
- p99
- 16.7s
- Throughput
- 31 tps
- Context
- 1.0M
- $/Mtok
- $0.200 / $0.400
phala - Together97.74%Healthy24h 96.47%
- p50
- 869ms
- p99
- 13.0s
- Throughput
- 48 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
together - Venice97.03%Healthy24h 96.03%
- p50
- 879ms
- p99
- 14.4s
- Throughput
- 64 tps
- Context
- 1.0M
- $/Mtok
- $0.175 / $0.350
venice - StreamLake95.71%Degraded24h 95.24%
- p50
- 2.58s
- p99
- 36.3s
- Throughput
- 40 tps
- Context
- 1.0M
- $/Mtok
- $0.079 / $0.157
streamlake/fp8fp8 - SiliconFlow95.65%Healthy24h 98.01%
- p50
- 1.93s
- p99
- 28.9s
- Throughput
- 64 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
siliconflow/fp8fp8 - Parasail95.53%Healthy24h 98.47%
- p50
- 756ms
- p99
- 13.5s
- Throughput
- 52 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
parasail/fp8fp8 - Mancer 293.02%Degraded24h 95.11%
- p50
- 1.09s
- p99
- 15.9s
- Throughput
- 31 tps
- Context
- 1.0M
- $/Mtok
- $0.145 / $0.450
mancer/fp8fp8 - AkashML85.39%Down24h 97.94%
- p50
- 807ms
- p99
- 9.66s
- Throughput
- 30 tps
- Context
- 131K
- $/Mtok
- $0.140 / $0.280
akashml/fp8fp8
Measured 1m 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 and p99 are OpenRouter's rolling 30-minute windows, averaged over the hour.
Throughput
Median tokens per second.