DeepSeek V4 Flash 0423: providers compared

Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.

Current state

	State
Cloudflarecloudflare	Healthy	100%◆ (best)	99.06%	99.72%	99.72%	4.65s	25.4s	50 tps	384K	$0.440 / $1.32
Novitanovita/fp8fp8	Healthy	100%◆ (best)	99.96%	99.95%	99.93%◆ (best)	1.34s	5.23s◆ (best)	63 tps	1.0M	$0.140 / $0.280
SiliconFlowsiliconflow/fp8fp8	Healthy	100%◆ (best)	99.41%	99.73%	98.28%	1.08s	7.00s	33 tps	1.0M	$0.130 / $0.280
GMICloudgmicloud/fp8fp8	Healthy	99.99%	99.93%	99.97%◆ (best)	99.69%	1.47s	5.57s	59 tps	1.0M	$0.091 / $0.182
StreamLakestreamlake/fp8fp8	Healthy	99.89%	98.93%	99.04%	97.67%	1.57s	24.4s	52 tps	1.0M	$0.042 / $0.084
Relacerelace/fp4fp4	Healthy	99.87%	99.92%	99.64%	99.73%	560ms◆ (best)	7.20s	93 tps	1.0M	$0.030 / $1.28
Baidubaidu/fp8fp8	Healthy	99.85%	99.58%	89.53%	96.37%	1.05s	9.19s	71 tps	1.0M	$0.140 / $0.280
DigitalOceandigitalocean	Healthy	99.83%	99.81%	99.82%	99.83%	1.01s	16.2s	16 tps	1.0M	$0.098 / $0.196
AtlasCloudatlas-cloud/fp4fp4	Healthy	99.73%	98.76%	98.47%	99.26%	1.83s	15.9s	50 tps	1.0M	$0.140 / $0.280
OpenInferenceopen-inference/fp4fp4	Healthy	99.59%	99.56%	98.63%	98.63%	826ms	8.64s	34 tps	1.0M	$0.039 / $1.66
Alibabaalibaba/fp8fp8	Healthy	99.01%	98.99%	98.78%	99.10%	856ms	6.15s	96 tps◆ (best)	1.0M	$0.134 / $0.268
Parasailparasail/fp8fp8	Healthy	98.07%	99.74%	99.80%	99.72%	959ms	11.6s	40 tps	1.0M	$0.140 / $0.280
Venicevenice	Degraded	94.00%	98.19%	98.88%	98.39%	1.93s	22.4s	27 tps	1.0M	$0.097 / $0.193
Mancer 2mancer/fp8fp8	Degraded	93.81%	96.47%	97.13%	97.91%	1.17s	20.0s	18 tps	1.0M	$0.190 / $0.500
Azureazure/us	Degraded	92.01%	96.81%	97.60%	95.91%	3.63s	49.5s	31 tps	1.0M	$0.210 / $0.560
CoreWeavecoreweave/fp8fp8	No data	—	—	—	—	—	—	—	1.0M	$0.140 / $0.280
Crusoecrusoe/fp8fp8	No data	—	—	—	—	—	—	—	1.0M	$0.140 / $0.280
DeepInfradeepinfra/fp8fp8	No data	—	99.81%	99.54%	99.49%	9.77s	25.1s	9.0 tps	1.0M	$0.090 / $0.180
DeepSeekdeepseek	No data	—	—	—	—	—	—	—	1.0M	$0.220 / $0.660
NextBitnextbit/fp8fp8	No data	—	—	—	99.34%	—	—	—	1.0M	$0.150 / $0.300
OpenInferenceopen-inference/fp8fp8	No data	—	—	95.43%	97.88%	—	—	—	1.0M	$0.002 / $1.31◆ (best)
Phalaphala	No data	—	—	—	99.71%	—	—	—	1.0M	$0.200 / $0.400
Rekareka	No data	—	—	98.07%	98.07%	—	—	—	1.0M	$0.130 / $0.260
Sail Researchsail-research/fp4fp4	No data	—	—	—	—	—	—	—	1.0M	$0.090 / $0.180
Waferwafer	No data	—	—	—	99.43%	—	—	—	1.0M	$0.100 / $0.250

Cloudflare100%◆ (best)
Healthy24h 99.06%
7d
99.72%
30d
99.72%
p50
4.65s
p99
25.4s
Throughput
50 tps
Context
384K
$/Mtok
$0.440 / $1.32
cloudflare
Novita100%◆ (best)
Healthy24h 99.96%
7d
99.95%
30d
99.93%◆ (best)
p50
1.34s
p99
5.23s◆ (best)
Throughput
63 tps
Context
1.0M
$/Mtok
$0.140 / $0.280
novita/fp8fp8
SiliconFlow100%◆ (best)
Healthy24h 99.41%
7d
99.73%
30d
98.28%
p50
1.08s
p99
7.00s
Throughput
33 tps
Context
1.0M
$/Mtok
$0.130 / $0.280
siliconflow/fp8fp8
GMICloud99.99%
Healthy24h 99.93%
7d
99.97%◆ (best)
30d
99.69%
p50
1.47s
p99
5.57s
Throughput
59 tps
Context
1.0M
$/Mtok
$0.091 / $0.182
gmicloud/fp8fp8
StreamLake99.89%
Healthy24h 98.93%
7d
99.04%
30d
97.67%
p50
1.57s
p99
24.4s
Throughput
52 tps
Context
1.0M
$/Mtok
$0.042 / $0.084
streamlake/fp8fp8
Relace99.87%
Healthy24h 99.92%
7d
99.64%
30d
99.73%
p50
560ms◆ (best)
p99
7.20s
Throughput
93 tps
Context
1.0M
$/Mtok
$0.030 / $1.28
relace/fp4fp4
Baidu99.85%
Healthy24h 99.58%
7d
89.53%
30d
96.37%
p50
1.05s
p99
9.19s
Throughput
71 tps
Context
1.0M
$/Mtok
$0.140 / $0.280
baidu/fp8fp8
DigitalOcean99.83%
Healthy24h 99.81%
7d
99.82%
30d
99.83%
p50
1.01s
p99
16.2s
Throughput
16 tps
Context
1.0M
$/Mtok
$0.098 / $0.196
digitalocean
AtlasCloud99.73%
Healthy24h 98.76%
7d
98.47%
30d
99.26%
p50
1.83s
p99
15.9s
Throughput
50 tps
Context
1.0M
$/Mtok
$0.140 / $0.280
atlas-cloud/fp4fp4
OpenInference99.59%
Healthy24h 99.56%
7d
98.63%
30d
98.63%
p50
826ms
p99
8.64s
Throughput
34 tps
Context
1.0M
$/Mtok
$0.039 / $1.66
open-inference/fp4fp4
Alibaba99.01%
Healthy24h 98.99%
7d
98.78%
30d
99.10%
p50
856ms
p99
6.15s
Throughput
96 tps◆ (best)
Context
1.0M
$/Mtok
$0.134 / $0.268
alibaba/fp8fp8
Parasail98.07%
Healthy24h 99.74%
7d
99.80%
30d
99.72%
p50
959ms
p99
11.6s
Throughput
40 tps
Context
1.0M
$/Mtok
$0.140 / $0.280
parasail/fp8fp8
Venice94.00%
Degraded24h 98.19%
7d
98.88%
30d
98.39%
p50
1.93s
p99
22.4s
Throughput
27 tps
Context
1.0M
$/Mtok
$0.097 / $0.193
venice
Mancer 293.81%
Degraded24h 96.47%
7d
97.13%
30d
97.91%
p50
1.17s
p99
20.0s
Throughput
18 tps
Context
1.0M
$/Mtok
$0.190 / $0.500
mancer/fp8fp8
Azure92.01%
Degraded24h 96.81%
7d
97.60%
30d
95.91%
p50
3.63s
p99
49.5s
Throughput
31 tps
Context
1.0M
$/Mtok
$0.210 / $0.560
azure/us
CoreWeave—
No data24h —
7d
—
30d
—
p50
—
p99
—
Throughput
—
Context
1.0M
$/Mtok
$0.140 / $0.280
coreweave/fp8fp8
Crusoe—
No data24h —
7d
—
30d
—
p50
—
p99
—
Throughput
—
Context
1.0M
$/Mtok
$0.140 / $0.280
crusoe/fp8fp8
DeepInfra—
No data24h 99.81%
7d
99.54%
30d
99.49%
p50
9.77s
p99
25.1s
Throughput
9.0 tps
Context
1.0M
$/Mtok
$0.090 / $0.180
deepinfra/fp8fp8
DeepSeek—
No data24h —
7d
—
30d
—
p50
—
p99
—
Throughput
—
Context
1.0M
$/Mtok
$0.220 / $0.660
deepseek
NextBit—
No data24h —
7d
—
30d
99.34%
p50
—
p99
—
Throughput
—
Context
1.0M
$/Mtok
$0.150 / $0.300
nextbit/fp8fp8
OpenInference—
No data24h —
7d
95.43%
30d
97.88%
p50
—
p99
—
Throughput
—
Context
1.0M
$/Mtok
$0.002 / $1.31◆ (best)
open-inference/fp8fp8
Phala—
No data24h —
7d
—
30d
99.71%
p50
—
p99
—
Throughput
—
Context
1.0M
$/Mtok
$0.200 / $0.400
phala
Reka—
No data24h —
7d
98.07%
30d
98.07%
p50
—
p99
—
Throughput
—
Context
1.0M
$/Mtok
$0.130 / $0.260
reka
Sail Research—
No data24h —
7d
—
30d
—
p50
—
p99
—
Throughput
—
Context
1.0M
$/Mtok
$0.090 / $0.180
sail-research/fp4fp4
Wafer—
No data24h —
7d
—
30d
99.43%
p50
—
p99
—
Throughput
—
Context
1.0M
$/Mtok
$0.100 / $0.250
wafer

Measured 5m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.

Price vs latency

One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag

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, averaged over the hour.

Throughput

Median tokens per second.