DeepSeek V4 Flash 0423

deepseek/deepseek-v4-flash

Compare with DeepSeek V4 Flash 0731, DeepSeek V4 Pro 0423, DeepSeek V4 Pro 0813

CompareDeepSeek V4 Flash 0423

	State
Cloudflarecloudflare	Healthy	100%	99.99%	809ms	4.90s	56 tps	384K	$0.440 / $1.32
DeepSeekdeepseek	Healthy	100%	99.89%	823ms	5.38s	79 tps	1.0M	$0.220 / $0.660
Novitanovita/fp8fp8	Healthy	100%	99.85%	1.07s	5.57s	71 tps	1.0M	$0.140 / $0.280
Baidubaidu/fp8fp8	Healthy	99.98%	98.50%	686ms	4.84s	83 tps	1.0M	$0.083 / $0.165
SiliconFlowsiliconflow/fp8fp8	Healthy	99.97%	99.35%	1.44s	8.82s	40 tps	1.0M	$0.130 / $0.280
AtlasCloudatlas-cloud/fp4fp4	Healthy	99.96%	99.91%	1.00s	4.31s	46 tps	1.0M	$0.140 / $0.280
Venicevenice	Healthy	99.96%	99.77%	1.61s	3.89s	34 tps	1.0M	$0.138 / $0.275
GMICloudgmicloud/fp8fp8	Healthy	99.95%	99.84%	2.27s	5.60s	48 tps	1.0M	$0.112 / $0.224
StreamLakestreamlake/fp8fp8	Healthy	99.92%	99.28%	2.81s	12.3s	37 tps	1.0M	$0.081 / $0.162
Alibabaalibaba/fp8fp8	Healthy	99.89%	99.81%	835ms	5.54s	87 tps	1.0M	$0.134 / $0.268
Sail Researchsail-research/fp4fp4	Healthy	99.66%	99.71%	763ms	6.66s	23 tps	1.0M	$0.090 / $0.180
DeepInfradeepinfra/fp8fp8	Healthy	99.61%	99.59%	775ms	4.20s	29 tps	1.0M	$0.090 / $0.180
CoreWeavecoreweave/fp8fp8	Healthy	99.47%	99.26%	520ms	9.48s	48 tps	1.0M	$0.140 / $0.280
Parasailparasail/fp8fp8	Healthy	99.30%	99.15%	708ms	9.38s	40 tps	1.0M	$0.140 / $0.280
Mancer 2mancer/fp8fp8	Healthy	99.19%	98.64%	1.00s	7.99s	33 tps	1.0M	$0.145 / $0.450
DigitalOceandigitalocean	Healthy	98.35%	99.08%	1.75s	21.5s	5.0 tps	1.0M	$0.068 / $0.168
Azureazure/us	Degraded	90.62%	85.04%	2.32s	30.1s	27 tps	1.0M	$0.210 / $0.560
Phalaphala	Down	—	96.28%	—	—	—	1.0M	$0.200 / $0.400

Cloudflare100%
Healthy24h 99.99%
p50
809ms
p99
4.90s
Throughput
56 tps
Context
384K
$/Mtok
$0.440 / $1.32
cloudflare
DeepSeek100%
Healthy24h 99.89%
p50
823ms
p99
5.38s
Throughput
79 tps
Context
1.0M
$/Mtok
$0.220 / $0.660
deepseek
Novita100%
Healthy24h 99.85%
p50
1.07s
p99
5.57s
Throughput
71 tps
Context
1.0M
$/Mtok
$0.140 / $0.280
novita/fp8fp8
Baidu99.98%
Healthy24h 98.50%
p50
686ms
p99
4.84s
Throughput
83 tps
Context
1.0M
$/Mtok
$0.083 / $0.165
baidu/fp8fp8
SiliconFlow99.97%
Healthy24h 99.35%
p50
1.44s
p99
8.82s
Throughput
40 tps
Context
1.0M
$/Mtok
$0.130 / $0.280
siliconflow/fp8fp8
AtlasCloud99.96%
Healthy24h 99.91%
p50
1.00s
p99
4.31s
Throughput
46 tps
Context
1.0M
$/Mtok
$0.140 / $0.280
atlas-cloud/fp4fp4
Venice99.96%
Healthy24h 99.77%
p50
1.61s
p99
3.89s
Throughput
34 tps
Context
1.0M
$/Mtok
$0.138 / $0.275
venice
GMICloud99.95%
Healthy24h 99.84%
p50
2.27s
p99
5.60s
Throughput
48 tps
Context
1.0M
$/Mtok
$0.112 / $0.224
gmicloud/fp8fp8
StreamLake99.92%
Healthy24h 99.28%
p50
2.81s
p99
12.3s
Throughput
37 tps
Context
1.0M
$/Mtok
$0.081 / $0.162
streamlake/fp8fp8
Alibaba99.89%
Healthy24h 99.81%
p50
835ms
p99
5.54s
Throughput
87 tps
Context
1.0M
$/Mtok
$0.134 / $0.268
alibaba/fp8fp8
Sail Research99.66%
Healthy24h 99.71%
p50
763ms
p99
6.66s
Throughput
23 tps
Context
1.0M
$/Mtok
$0.090 / $0.180
sail-research/fp4fp4
DeepInfra99.61%
Healthy24h 99.59%
p50
775ms
p99
4.20s
Throughput
29 tps
Context
1.0M
$/Mtok
$0.090 / $0.180
deepinfra/fp8fp8
CoreWeave99.47%
Healthy24h 99.26%
p50
520ms
p99
9.48s
Throughput
48 tps
Context
1.0M
$/Mtok
$0.140 / $0.280
coreweave/fp8fp8
Parasail99.30%
Healthy24h 99.15%
p50
708ms
p99
9.38s
Throughput
40 tps
Context
1.0M
$/Mtok
$0.140 / $0.280
parasail/fp8fp8
Mancer 299.19%
Healthy24h 98.64%
p50
1.00s
p99
7.99s
Throughput
33 tps
Context
1.0M
$/Mtok
$0.145 / $0.450
mancer/fp8fp8
DigitalOcean98.35%
Healthy24h 99.08%
p50
1.75s
p99
21.5s
Throughput
5.0 tps
Context
1.0M
$/Mtok
$0.068 / $0.168
digitalocean
Azure90.62%
Degraded24h 85.04%
p50
2.32s
p99
30.1s
Throughput
27 tps
Context
1.0M
$/Mtok
$0.210 / $0.560
azure/us
Phala—
Down24h 96.28%
p50
—
p99
—
Throughput
—
Context
1.0M
$/Mtok
$0.200 / $0.400
phala

Measured 4m 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.