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
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100% | 99.49% | 99.52% | 99.52% | 1.58s | 13.6s | 87 tps | 1.0M | $0.440 / $1.32 | |
| Healthy | 100% | 96.03% | 97.78% | 97.78% | 441ms | 4.28s | 64 tps | 1.0M | $0.130 / $0.260 | |
Cloudflarecloudflare | Healthy | 100% | 96.74% | 98.04% | 98.04% | 917ms | 11.5s | 55 tps | 1.3M | $0.440 / $1.32 |
DeepSeekdeepseek | Healthy | 100% | 98.94% | 83.69% | 83.69% | 791ms | 3.56s | 81 tps | 1.0M | $0.440 / $1.32 |
| Healthy | 99.99% | 99.42% | 99.36% | 99.36% | 1.25s | 8.75s | 99 tps | 1.0M | $0.140 / $0.280 | |
| Healthy | 99.95% | 99.29% | 97.37% | 97.37% | 905ms | 10.5s | 15 tps | 1.0M | $0.080 / $0.180 | |
| Healthy | 99.94% | 99.63% | 99.80% | 99.80% | 979ms | 9.73s | 135 tps | 262K | $0.130 / $0.280 | |
| Healthy | 99.91% | 99.62% | 99.64% | 99.64% | 889ms | 6.05s | 64 tps | 1.0M | $0.070 / $0.140 | |
| Healthy | 99.87% | 99.52% | 99.00% | 99.00% | 1.55s | 7.80s | 161 tps | 262K | $0.220 / $0.660 | |
| Healthy | 99.81% | 99.52% | 98.61% | 98.61% | 1.67s | 25.7s | 13 tps | 1.0M | $0.120 / $0.278 | |
| Healthy | 99.80% | 99.75% | 99.16% | 99.16% | 674ms | 6.34s | 51 tps | 1.0M | $0.130 / $0.280 | |
| Down | 99.76% | 88.54% | 79.32% | 79.32% | 4.16s | 28.7s | 33 tps | 262K | $0.077 / $0.155 | |
Alibabaalibaba | Healthy | 99.64% | 99.15% | 98.72% | 98.72% | 1.93s | 13.6s | 63 tps | 1.0M | $0.402 / $1.21 |
| Healthy | 99.50% | 97.79% | 98.57% | 98.57% | 3.48s | 59.5s | 15 tps | 1.0M | $0.065 / $0.180 | |
| Healthy | 99.46% | 98.85% | 99.19% | 99.19% | 2.44s | 12.7s | 62 tps | 1.0M | $0.084 / $0.168 | |
| Healthy | 99.37% | 98.91% | 97.32% | 97.32% | 1.94s | 59.6s | 34 tps | 1.0M | $0.079 / $0.157 | |
| Healthy | 99.29% | 98.93% | 98.55% | 98.55% | 667ms | 11.6s | 50 tps | 1.0M | $0.080 / $0.180 | |
Phalaphala | Healthy | 98.92% | 98.46% | 97.82% | 97.82% | 883ms | 18.1s | 39 tps | 1.0M | $0.200 / $0.400 |
| Healthy | 98.75% | 99.24% | 99.45% | 99.45% | 895ms | 8.65s | 96 tps | 1.0M | $0.140 / $0.280 | |
| Healthy | 98.71% | 98.59% | 98.75% | 98.75% | 1.80s | 13.0s | 50 tps | 1.0M | $0.140 / $0.280 | |
DigitalOceandigitalocean | Healthy | 98.59% | 97.32% | 95.85% | 95.85% | 915ms | 7.69s | 10 tps | 1.0M | $0.080 / $0.252 |
Togethertogether | Degraded | 98.50% | 97.32% | 97.28% | 97.28% | 1.04s | 14.4s | 44 tps | 1.0M | $0.140 / $0.280 |
| Healthy | 98.49% | 98.25% | 98.20% | 98.20% | 953ms | 20.2s | 52 tps | 1.0M | $0.140 / $0.280 | |
Fireworksfireworks | Healthy | 97.67% | 98.79% | 97.42% | 97.42% | 814ms | 22.2s | 69 tps | 1.0M | $0.140 / $0.280 |
Venicevenice | Healthy | 97.53% | 97.38% | 97.47% | 97.47% | 926ms | 13.3s | 64 tps | 1.0M | $0.175 / $0.350 |
| Healthy | 95.00% | 95.51% | 95.45% | 95.45% | 911ms | 12.0s | 54 tps | 1.0M | $0.145 / $0.450 | |
| Degraded | 94.70% | 98.23% | 97.90% | 97.90% | 1.51s | 16.2s | 28 tps | 131K | $0.140 / $0.280 | |
Waferwafer/fast | Degraded | 94.43% | 98.39% | 98.61% | 98.61% | 894ms | 11.1s | 194 tps | 1.0M | $0.280 / $0.560 |
| Degraded | 92.96% | 96.96% | 97.75% | 97.75% | 939ms | 20.5s | 62 tps | 262K | $0.189 / $0.420 | |
| Degraded | 92.33% | 93.96% | 94.83% | 94.83% | 3.32s | 28.3s | 9.0 tps | 262K | $0.065 / $0.180 |
- AtlasCloud100%Healthy24h 99.49%
- 7d
- 99.52%
- 30d
- 99.52%
- p50
- 1.58s
- p99
- 13.6s
- Throughput
- 87 tps
- Context
- 1.0M
- $/Mtok
- $0.440 / $1.32
atlas-cloud/fp4fp4 - BaseTen100%Healthy24h 96.03%
- 7d
- 97.78%
- 30d
- 97.78%
- p50
- 441ms
- p99
- 4.28s
- Throughput
- 64 tps
- Context
- 1.0M
- $/Mtok
- $0.130 / $0.260
baseten/fp8fp8 - Cloudflare100%Healthy24h 96.74%
- 7d
- 98.04%
- 30d
- 98.04%
- p50
- 917ms
- p99
- 11.5s
- Throughput
- 55 tps
- Context
- 1.3M
- $/Mtok
- $0.440 / $1.32
cloudflare - DeepSeek100%Healthy24h 98.94%
- 7d
- 83.69%
- 30d
- 83.69%
- p50
- 791ms
- p99
- 3.56s
- Throughput
- 81 tps
- Context
- 1.0M
- $/Mtok
- $0.440 / $1.32
deepseek - Novita99.99%Healthy24h 99.42%
- 7d
- 99.36%
- 30d
- 99.36%
- p50
- 1.25s
- p99
- 8.75s
- Throughput
- 99 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
novita/fp8fp8 - Ambient99.95%Healthy24h 99.29%
- 7d
- 97.37%
- 30d
- 97.37%
- p50
- 905ms
- p99
- 10.5s
- Throughput
- 15 tps
- Context
- 1.0M
- $/Mtok
- $0.080 / $0.180
ambient/fp4fp4 - CoreWeave99.94%Healthy24h 99.63%
- 7d
- 99.80%
- 30d
- 99.80%
- p50
- 979ms
- p99
- 9.73s
- Throughput
- 135 tps
- Context
- 262K
- $/Mtok
- $0.130 / $0.280
coreweave/fp8fp8 - Relace99.91%Healthy24h 99.62%
- 7d
- 99.64%
- 30d
- 99.64%
- p50
- 889ms
- p99
- 6.05s
- Throughput
- 64 tps
- Context
- 1.0M
- $/Mtok
- $0.070 / $0.140
relace/fp4fp4 - Reka99.87%Healthy24h 99.52%
- 7d
- 99.00%
- 30d
- 99.00%
- p50
- 1.55s
- p99
- 7.80s
- Throughput
- 161 tps
- Context
- 262K
- $/Mtok
- $0.220 / $0.660
reka/fp4fp4 - Morph99.81%Healthy24h 99.52%
- 7d
- 98.61%
- 30d
- 98.61%
- p50
- 1.67s
- p99
- 25.7s
- Throughput
- 13 tps
- Context
- 1.0M
- $/Mtok
- $0.120 / $0.278
morph/bf16bf16 - Inceptron99.80%Healthy24h 99.75%
- 7d
- 99.16%
- 30d
- 99.16%
- p50
- 674ms
- p99
- 6.34s
- Throughput
- 51 tps
- Context
- 1.0M
- $/Mtok
- $0.130 / $0.280
inceptron/fp4fp4 - Decart99.76%Down24h 88.54%
- 7d
- 79.32%
- 30d
- 79.32%
- p50
- 4.16s
- p99
- 28.7s
- Throughput
- 33 tps
- Context
- 262K
- $/Mtok
- $0.077 / $0.155
decart/fp4fp4 - Alibaba99.64%Healthy24h 99.15%
- 7d
- 98.72%
- 30d
- 98.72%
- p50
- 1.93s
- p99
- 13.6s
- Throughput
- 63 tps
- Context
- 1.0M
- $/Mtok
- $0.402 / $1.21
alibaba - Sail Research99.50%Healthy24h 97.79%
- 7d
- 98.57%
- 30d
- 98.57%
- p50
- 3.48s
- p99
- 59.5s
- Throughput
- 15 tps
- Context
- 1.0M
- $/Mtok
- $0.065 / $0.180
sail-research/fp4fp4 - GMICloud99.46%Healthy24h 98.85%
- 7d
- 99.19%
- 30d
- 99.19%
- p50
- 2.44s
- p99
- 12.7s
- Throughput
- 62 tps
- Context
- 1.0M
- $/Mtok
- $0.084 / $0.168
gmicloud/fp8fp8 - StreamLake99.37%Healthy24h 98.91%
- 7d
- 97.32%
- 30d
- 97.32%
- p50
- 1.94s
- p99
- 59.6s
- Throughput
- 34 tps
- Context
- 1.0M
- $/Mtok
- $0.079 / $0.157
streamlake/fp8fp8 - DeepInfra99.29%Healthy24h 98.93%
- 7d
- 98.55%
- 30d
- 98.55%
- p50
- 667ms
- p99
- 11.6s
- Throughput
- 50 tps
- Context
- 1.0M
- $/Mtok
- $0.080 / $0.180
deepinfra/fp8fp8 - Phala98.92%Healthy24h 98.46%
- 7d
- 97.82%
- 30d
- 97.82%
- p50
- 883ms
- p99
- 18.1s
- Throughput
- 39 tps
- Context
- 1.0M
- $/Mtok
- $0.200 / $0.400
phala - Baidu98.75%Healthy24h 99.24%
- 7d
- 99.45%
- 30d
- 99.45%
- p50
- 895ms
- p99
- 8.65s
- Throughput
- 96 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
baidu/fp8fp8 - SiliconFlow98.71%Healthy24h 98.59%
- 7d
- 98.75%
- 30d
- 98.75%
- p50
- 1.80s
- p99
- 13.0s
- Throughput
- 50 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
siliconflow/fp8fp8 - DigitalOcean98.59%Healthy24h 97.32%
- 7d
- 95.85%
- 30d
- 95.85%
- p50
- 915ms
- p99
- 7.69s
- Throughput
- 10 tps
- Context
- 1.0M
- $/Mtok
- $0.080 / $0.252
digitalocean - Together98.50%Degraded24h 97.32%
- 7d
- 97.28%
- 30d
- 97.28%
- p50
- 1.04s
- p99
- 14.4s
- Throughput
- 44 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
together - Parasail98.49%Healthy24h 98.25%
- 7d
- 98.20%
- 30d
- 98.20%
- p50
- 953ms
- p99
- 20.2s
- Throughput
- 52 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
parasail/fp8fp8 - Fireworks97.67%Healthy24h 98.79%
- 7d
- 97.42%
- 30d
- 97.42%
- p50
- 814ms
- p99
- 22.2s
- Throughput
- 69 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
fireworks - Venice97.53%Healthy24h 97.38%
- 7d
- 97.47%
- 30d
- 97.47%
- p50
- 926ms
- p99
- 13.3s
- Throughput
- 64 tps
- Context
- 1.0M
- $/Mtok
- $0.175 / $0.350
venice - Mancer 295.00%Healthy24h 95.51%
- 7d
- 95.45%
- 30d
- 95.45%
- p50
- 911ms
- p99
- 12.0s
- Throughput
- 54 tps
- Context
- 1.0M
- $/Mtok
- $0.145 / $0.450
mancer/fp8fp8 - AkashML94.70%Degraded24h 98.23%
- 7d
- 97.90%
- 30d
- 97.90%
- p50
- 1.51s
- p99
- 16.2s
- Throughput
- 28 tps
- Context
- 131K
- $/Mtok
- $0.140 / $0.280
akashml/fp8fp8 - Wafer94.43%Degraded24h 98.39%
- 7d
- 98.61%
- 30d
- 98.61%
- p50
- 894ms
- p99
- 11.1s
- Throughput
- 194 tps
- Context
- 1.0M
- $/Mtok
- $0.280 / $0.560
wafer/fast - Io Net92.96%Degraded24h 96.96%
- 7d
- 97.75%
- 30d
- 97.75%
- p50
- 939ms
- p99
- 20.5s
- Throughput
- 62 tps
- Context
- 262K
- $/Mtok
- $0.189 / $0.420
io-net/fp8fp8 - OpenInference92.33%Degraded24h 93.96%
- 7d
- 94.83%
- 30d
- 94.83%
- p50
- 3.32s
- p99
- 28.3s
- Throughput
- 9.0 tps
- Context
- 262K
- $/Mtok
- $0.065 / $0.180
open-inference/fp4fp4
Measured 1m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
Price changes
- DeepSeek V4 Flash 0731OpenInferenceopen-inference/fp4$0.060 / $0.180 → $0.065 / $0.1802026-08-21 08:05 UTC
- DeepSeek V4 Flash 0731Morphmorph/bf16$0.120 / $0.278 → $0.120 / $0.2782026-08-21 08:05 UTC
- DeepSeek V4 Flash 0731OpenInferenceopen-inference/fp4$0.080 / $0.180 → $0.060 / $0.1802026-08-21 07:55 UTC
- DeepSeek V4 Flash 0731OpenInferenceopen-inference/fp4$0.065 / $0.140 → $0.080 / $0.1802026-08-21 06:25 UTC
- DeepSeek V4 Flash 0731DeepSeekdeepseek$0.220 / $0.660 → $0.440 / $1.322026-08-21 06:06 UTC
- DeepSeek V4 Flash 0731Fireworksfireworks$0.220 / $0.660 → $0.140 / $0.2802026-08-21 05:25 UTC
- DeepSeek V4 Flash 0731Fireworksfireworks$0.140 / $0.280 → $0.220 / $0.6602026-08-21 04:50 UTC
- DeepSeek V4 Flash 0731DeepSeekdeepseek$0.440 / $1.32 → $0.220 / $0.6602026-08-21 04:10 UTC
- DeepSeek V4 Flash 0731DeepSeekdeepseek$0.220 / $0.660 → $0.440 / $1.322026-08-21 01:10 UTC
- DeepSeek V4 Flash 0731Decartdecart/fp4$0.066 / $0.131 → $0.077 / $0.1552026-08-21 00:50 UTC
- DeepSeek V4 Flash 0731Morphmorph/bf16$0.139 / $0.278 → $0.120 / $0.2782026-08-21 00:45 UTC
- DeepSeek V4 Flash 0731AtlasCloudatlas-cloud/fp4$0.140 / $0.280 → $0.440 / $1.322026-08-21 00:10 UTC
- DeepSeek V4 Flash 0731Alibabaalibaba$0.201 / $0.603 → $0.402 / $1.212026-08-21 00:05 UTC
- DeepSeek V4 Flash 0731StreamLakestreamlake/fp8$0.067 / $0.134 → $0.079 / $0.1572026-08-20 23:50 UTC
- DeepSeek V4 Flash 0731Baidubaidu/fp8$0.069 / $0.137 → $0.140 / $0.2802026-08-20 23:40 UTC
- DeepSeek V4 Flash 0731Decartdecart/fp4$0.076 / $0.153 → $0.066 / $0.1312026-08-20 19:55 UTC
- DeepSeek V4 Flash 0731Ambientambient/fp4$0.800 / $0.180 → $0.080 / $0.1802026-08-20 19:30 UTC
- DeepSeek V4 Flash 0731StreamLakestreamlake/fp8$0.078 / $0.157 → $0.067 / $0.1342026-08-20 19:05 UTC
- DeepSeek V4 Flash 0731Ambientambient/fp4$0.140 / $0.280 → $0.800 / $0.1802026-08-20 18:50 UTC
- DeepSeek V4 Flash 0731GMICloudgmicloud/fp8$0.112 / $0.224 → $0.084 / $0.1682026-08-20 18:45 UTC
- DeepSeek V4 Flash 0731Baidubaidu/fp8$0.080 / $0.160 → $0.069 / $0.1372026-08-20 18:45 UTC
- DeepSeek V4 Flash 0731Alibabaalibaba$0.242 / $0.726 → $0.201 / $0.6032026-08-20 16:35 UTC
- DeepSeek V4 Flash 0731Alibabaalibaba$0.484 / $1.45 → $0.242 / $0.7262026-08-20 14:10 UTC
- DeepSeek V4 Flash 0731StreamLakestreamlake/fp8$0.079 / $0.157 → $0.078 / $0.1572026-08-20 13:05 UTC
- DeepSeek V4 Flash 0731Baidubaidu/fp8$0.140 / $0.280 → $0.080 / $0.1602026-08-20 12:50 UTC
- DeepSeek V4 Flash 0731DeepSeekdeepseek$0.440 / $1.32 → $0.220 / $0.6602026-08-20 10:10 UTC
- DeepSeek V4 Flash 0731Io Netio-net/fp8$0.149 / $0.320 → $0.189 / $0.4202026-08-20 09:25 UTC
- DeepSeek V4 Flash 0731Rekareka/fp4$0.220 / $0.660 → $0.220 / $0.6602026-08-20 08:05 UTC
- DeepSeek V4 Flash 0731DeepSeekdeepseek$0.220 / $0.660 → $0.440 / $1.322026-08-20 06:10 UTC
- DeepSeek V4 Flash 0731Cloudflarecloudflare$0.080 / $0.180 → $0.440 / $1.322026-08-20 05:50 UTC
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.