Replicate
Replicate.com is a mature platform for deploying and sharing AI models.
Replicate Serverless GPU Prices
CPU | 0 | 8 | 4 | 0.0001/s 0.36/h 0 | |
| 16 | 8 | 4 | 0.000225/s 0.81/h 0 | ||
| 48 | 16 | 4 | 0.000575/s 2.07/h -12% | ||
| 48 | 72 | 10 | 0.000725/s 2.61/h +12% | ||
| 40 | 72 | 10 | 0.00115/s 4.14/h -10% | ||
| 80 | 144 | 10 | 0.0014/s 5.04/h +10% | ||
| 384 | 680 | 48 | 0.0058/s 20.88/h |
Recent benchmarks
37.42s 12:01:26 AM
17.31s 12:01:48 AM
65.29s 12:02:27 AM
11.35s 12:01:52 AM
12.35s 12:01:57 AM
24.01s 12:02:37 AM
142.7s 7:49:42 PM
13.18s 12:02:05 AM
20.57s 12:02:10 AM
13.46s 12:03:53 AM
114.63s 12:04:37 AM
91.68s 7:48:54 PM
20.85s 12:02:48 AM
13.22s 12:01:27 AM
106.6s 7:48:52 PM
18.54s 12:01:32 AM
15.83s 12:02:27 AM
95.46s 7:48:51 PM
13.22s 12:02:12 AM
12.23s 12:01:59 AM
96.15s 7:48:45 PM
10.07s 12:02:40 AM
236.94s 12:05:30 AM
85.66s 7:48:49 PM
12.28s 12:01:36 AM
18.23s 12:01:47 AM
95.11s 7:49:08 PM
10.89s 12:01:46 AM
13.2s 12:02:21 AM
89.23s 7:48:48 PM
12.87s 12:01:50 AM
21.95s 12:01:53 AM
hello_torch by replicate on t4
Median84.99s
Average105.28s
Failures0
80.6s 12:01:26 AM
77.8s 12:01:48 AM
87.9s 12:02:27 AM
85.39s 12:01:52 AM
82.3s 12:01:57 AM
95.89s 12:02:37 AM
118.41s 7:49:42 PM
89.87s 12:02:05 AM
95.66s 12:02:10 AM
227.53s 12:03:53 AM
246.83s 12:04:37 AM
76.33s 7:48:54 PM
154.69s 12:02:48 AM
78.88s 12:01:27 AM
95.61s 7:48:52 PM
84.58s 12:01:32 AM
88.33s 12:02:27 AM
81.1s 7:48:51 PM
97.4s 12:02:12 AM
78.99s 12:01:59 AM
82.31s 7:48:45 PM
99.99s 12:02:40 AM
320.89s 12:05:30 AM
74.02s 7:48:49 PM
75.92s 12:01:36 AM
75.59s 12:01:47 AM
83.52s 7:49:08 PM
81.07s 12:01:46 AM
84.18s 12:02:21 AM
78.25s 7:48:48 PM
92.84s 12:01:50 AM
96.27s 12:01:53 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.