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
Fireworksfireworks	Healthy	100%	98.36%	99.23%	99.23%	1.92s	22.6s	52 tps	1.0M	$1.32 / $3.96
SiliconFlowsiliconflow/fp8fp8	Healthy	100%	98.51%	99.03%	99.03%	1.46s	61.9s	55 tps	1.0M	$1.32 / $3.96
StreamLakestreamlake	Healthy	100%	98.21%	98.76%	98.76%	1.55s	8.26s	54 tps	1.0M	$1.32 / $3.96
DeepSeekdeepseek	Healthy	99.93%	99.95%	85.10%	85.10%	1.29s	8.45s	38 tps	1.0M	$1.32 / $3.96
Novitanovita/fp8fp8	Healthy	99.89%	99.59%	98.91%	98.91%	1.23s	4.34s	43 tps	1.0M	$1.32 / $3.96
GMICloudgmicloud/fp8fp8	Healthy	99.85%	99.58%	99.64%	99.64%	4.42s	26.3s	41 tps	1.0M	$1.19 / $3.56
Cloudflarecloudflare	Degraded	98.43%	97.53%	97.16%	97.16%	4.40s	58.9s	42 tps	1.0M	$1.32 / $3.96
Alibabaalibaba	Healthy	97.86%	99.85%	99.62%	99.62%	1.29s	6.76s	48 tps	1.0M	$1.21 / $3.62
Togethertogether	Healthy	95.96%	96.11%	96.91%	96.91%	1.86s	43.6s	60 tps	1.0M	$1.32 / $3.96
Parasailparasail/fp8fp8	Degraded	94.07%	94.94%	95.62%	95.62%	1.30s	42.1s	46 tps	1.0M	$1.32 / $3.96
DigitalOceandigitalocean	Degraded	93.62%	97.80%	96.99%	96.99%	1.53s	18.2s	37 tps	1.0M	$1.32 / $3.96
BaseTenbaseten/fp4fp4	No data	—	97.56%	98.54%	98.54%	1.85s	28.8s	101 tps	1.0M	$1.32 / $3.96

Fireworks100%
Healthy24h 98.36%
7d
99.23%
30d
99.23%
p50
1.92s
p99
22.6s
Throughput
52 tps
Context
1.0M
$/Mtok
$1.32 / $3.96
fireworks
SiliconFlow100%
Healthy24h 98.51%
7d
99.03%
30d
99.03%
p50
1.46s
p99
61.9s
Throughput
55 tps
Context
1.0M
$/Mtok
$1.32 / $3.96
siliconflow/fp8fp8
StreamLake100%
Healthy24h 98.21%
7d
98.76%
30d
98.76%
p50
1.55s
p99
8.26s
Throughput
54 tps
Context
1.0M
$/Mtok
$1.32 / $3.96
streamlake
DeepSeek99.93%
Healthy24h 99.95%
7d
85.10%
30d
85.10%
p50
1.29s
p99
8.45s
Throughput
38 tps
Context
1.0M
$/Mtok
$1.32 / $3.96
deepseek
Novita99.89%
Healthy24h 99.59%
7d
98.91%
30d
98.91%
p50
1.23s
p99
4.34s
Throughput
43 tps
Context
1.0M
$/Mtok
$1.32 / $3.96
novita/fp8fp8
GMICloud99.85%
Healthy24h 99.58%
7d
99.64%
30d
99.64%
p50
4.42s
p99
26.3s
Throughput
41 tps
Context
1.0M
$/Mtok
$1.19 / $3.56
gmicloud/fp8fp8
Cloudflare98.43%
Degraded24h 97.53%
7d
97.16%
30d
97.16%
p50
4.40s
p99
58.9s
Throughput
42 tps
Context
1.0M
$/Mtok
$1.32 / $3.96
cloudflare
Alibaba97.86%
Healthy24h 99.85%
7d
99.62%
30d
99.62%
p50
1.29s
p99
6.76s
Throughput
48 tps
Context
1.0M
$/Mtok
$1.21 / $3.62
alibaba
Together95.96%
Healthy24h 96.11%
7d
96.91%
30d
96.91%
p50
1.86s
p99
43.6s
Throughput
60 tps
Context
1.0M
$/Mtok
$1.32 / $3.96
together
Parasail94.07%
Degraded24h 94.94%
7d
95.62%
30d
95.62%
p50
1.30s
p99
42.1s
Throughput
46 tps
Context
1.0M
$/Mtok
$1.32 / $3.96
parasail/fp8fp8
DigitalOcean93.62%
Degraded24h 97.80%
7d
96.99%
30d
96.99%
p50
1.53s
p99
18.2s
Throughput
37 tps
Context
1.0M
$/Mtok
$1.32 / $3.96
digitalocean
BaseTen—
No data24h 97.56%
7d
98.54%
30d
98.54%
p50
1.85s
p99
28.8s
Throughput
101 tps
Context
1.0M
$/Mtok
$1.32 / $3.96
baseten/fp4fp4

Measured 3m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.

Price changes

DeepSeek V4 Pro 0813 DeepSeekdeepseek$0.660 / $1.98 → $1.32 / $3.962026-08-21 06:06 UTC
DeepSeek V4 Pro 0813 DeepSeekdeepseek$1.32 / $3.96 → $0.660 / $1.982026-08-21 04:10 UTC
DeepSeek V4 Pro 0813 DeepSeekdeepseek$0.660 / $1.98 → $1.32 / $3.962026-08-21 01:10 UTC
DeepSeek V4 Pro 0813 Alibabaalibaba$0.603 / $1.81 → $1.21 / $3.622026-08-21 00:10 UTC
DeepSeek V4 Pro 0813 Alibabaalibaba$0.726 / $2.18 → $0.603 / $1.812026-08-20 16:35 UTC
DeepSeek V4 Pro 0813 Alibabaalibaba$1.32 / $3.96 → $0.726 / $2.182026-08-20 14:05 UTC
DeepSeek V4 Pro 0813 DeepSeekdeepseek$1.32 / $3.96 → $0.660 / $1.982026-08-20 10:10 UTC
DeepSeek V4 Pro 0813 DeepSeekdeepseek$0.660 / $1.98 → $1.32 / $3.962026-08-20 06:05 UTC
DeepSeek V4 Pro 0813 DeepSeekdeepseek$1.32 / $3.96 → $0.660 / $1.982026-08-20 04:05 UTC
DeepSeek V4 Pro 0813 GMICloudgmicloud/fp8$1.06 / $3.17 → $1.19 / $3.562026-08-20 01:20 UTC
DeepSeek V4 Pro 0813 DeepSeekdeepseek$0.660 / $1.98 → $1.32 / $3.962026-08-20 01:05 UTC
DeepSeek V4 Pro 0813 Alibabaalibaba$0.726 / $2.18 → $1.32 / $3.962026-08-20 00:05 UTC
DeepSeek V4 Pro 0813 GMICloudgmicloud/fp8$1.19 / $3.56 → $1.06 / $3.172026-08-19 19:55 UTC
DeepSeek V4 Pro 0813 Alibabaalibaba$1.32 / $3.96 → $0.726 / $2.182026-08-19 14:05 UTC
DeepSeek V4 Pro 0813 DeepSeekdeepseek$1.32 / $3.96 → $0.660 / $1.982026-08-19 10:10 UTC
DeepSeek V4 Pro 0813 DeepSeekdeepseek$0.660 / $1.98 → $1.32 / $3.962026-08-19 06:10 UTC
DeepSeek V4 Pro 0813 DeepSeekdeepseek$1.32 / $3.96 → $0.660 / $1.982026-08-19 04:10 UTC

History

1 endpoint reporting no data is 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 is OpenRouter's rolling 30-minute window.

Throughput

Median tokens per second.