Replicate
Replicate.com is a mature platform for deploying and sharing AI models.

Replicate Serverless GPU Prices

CPU
084
0.0001/s
0.36/h
0
1684
0.000225/s
0.81/h
0
48164
0.000575/s
2.07/h
-12%
487210
0.000725/s
2.61/h
+12%
407210
0.00115/s
4.14/h
-10%
8014410
0.0014/s
5.04/h
+10%
38468048
0.0058/s
20.88/h

Recent benchmarks

hello_gpu by replicate on t4
Median15.01s
Average30.76s
Failures0
189.38s 12:03:53 AM
22.06s 12:02:20 AM
11.97s 12:02:25 AM
33.48s 12:03:15 AM
10.58s 12:01:42 AM
24.02s 12:02:06 AM
12.59s 12:02:20 AM
11.39s 12:02:25 AM
17.5s 12:05:20 AM
39.28s 12:01:33 AM
42.09s 12:02:26 AM
20.96s 12:02:21 AM
11.09s 12:01:29 AM
15.44s 12:01:46 AM
47.75s 12:04:21 AM
20.05s 12:02:06 AM
87.29s 12:02:11 AM
12.1s 12:01:37 AM
17.84s 12:02:07 AM
30.64s 12:02:07 AM
12.25s 12:01:40 AM
9.82s 12:02:16 AM
12.04s 12:01:43 AM
14.58s 12:01:23 AM
14.01s 12:02:38 AM
12.06s 12:02:16 AM
21.8s 12:04:40 AM
12.45s 12:02:03 AM
160.09s 12:03:29 AM
11.5s 12:01:50 AM
13.43s 12:01:47 AM
12.94s 12:02:15 AM
hello_torch by replicate on t4
Median88.96s
Average108.66s
Failures0
82.81s 12:03:53 AM
81.05s 12:02:20 AM
88.72s 12:02:25 AM
144.93s 12:03:15 AM
84.11s 12:01:42 AM
111.26s 12:02:06 AM
95.92s 12:02:20 AM
83.89s 12:02:25 AM
298.31s 12:05:20 AM
82.58s 12:01:33 AM
129.59s 12:02:26 AM
81.41s 12:02:21 AM
86.33s 12:01:29 AM
86.62s 12:01:46 AM
233.06s 12:04:21 AM
90.83s 12:02:06 AM
98.81s 12:02:11 AM
81.92s 12:01:37 AM
83.07s 12:02:07 AM
106.56s 12:02:07 AM
89.2s 12:01:40 AM
92.39s 12:02:16 AM
88.43s 12:01:43 AM
78.54s 12:01:23 AM
104.44s 12:02:38 AM
82.06s 12:02:16 AM
219.8s 12:04:40 AM
82.42s 12:02:03 AM
146.07s 12:03:29 AM
76.69s 12:01:50 AM
95.22s 12:01:47 AM
90.06s 12:02:15 AM

About Replicate

There are large amounts of models made by the community as a showcase and for demonstration on the platform. You can publish your own model on Replicate for free, but you pay per second used by server.

With public models you only pay for the time the server is running, and with private models you also pay for the start up time.

You develop functions using the COG SDK, which makes it quite easy to containerize and deploy your AI models.

Replicate has stable support for webhooks, logging, and offer some observability.

The start up time for functions on Replicate is quite long (as of March 2024), so if fast responses are needed, it's better to use another vendor or configure Replicate to always keep a replica online.