Replicate 🇺🇸
Run AI/ML models without complex infrastructure
Founded in 2019, Replicate is a platform to train and deploy ML models in the cloud. You can run and fine tune open source models such as Meta's Llama 3, Mistral and Stable Diffusion without having to set up complex infrastructure.
They also feature a collection of generative models shared by the community, which you can run via the web UI or using their REST API.
Example customers include BuzzFeed, PhotoAI, Magnific, Unsplash, HeadshotPro.
What's good about Replicate
- Run AI/ML models without worrying about infrastructure and scaling
- Excellent web UI to explore and try out models without code
- Train and deploy custom models using Cog
Replicate's Pricing
Replicate uses a metered billing model. You only pay as long as your code is running, billed by the second based on the hardware selected.
Cloud GPUs are available in the following configurations:
| Name | GPUs | Total GPU Memory | vCPUs | RAM | Price per hour |
|---|---|---|---|---|---|
| Nvidia T4 | 1x T4 | 16GB | 4 | 16GB | $0.81 |
| Nvidia A40 (Small) | 1x A40 | 48GB | 4 | 16GB | $2.07 |
| Nvidia A40 (Large) | 1x A40 | 48GB | 10 | 72GB | $2.61 |
| Nvidia A100 (40GB) | 1x A100 | 40GB | 10 | 72GB | $4.14 |
| Nvidia A100 (80GB) | 1x A100 | 80GB | 10 | 144GB | $5.04 |
| 2x Nvidia A40 (Large) | 2x A40 | 96GB | 20 | 144GB | $5.22 |
| 2x Nvidia A100 (40GB) | 2x A100 | 80GB | 20 | 144GB | $8.28 |
| 2x Nvidia A100 (80GB) | 2x A100 | 160GB | 20 | 288GB | $10.08 |
| 4x Nvidia A40 (Large) | 4x A40 | 192GB | 40 | 288GB | $10.44 |
| 4x Nvidia A100 (40GB) | 4x A100 | 160GB | 40 | 288GB | $16.56 |
| 4x Nvidia A100 (80GB) | 4x A100 | 320GB | 40 | 576GB | $20.16 |
| 8x Nvidia A40 (Large) | 8x A40 | 384GB | 48 | 680GB | $20.88 |
| 8x Nvidia A100 (80GB) | 8x A100 | 640GB | 80 | 960GB | $40.32 |
Note: Our pricing examples are based on several assumptions. Your actual costs may differ. Always check the cloud provider's website for the most up-to-date pricing. You can find Replicate's latest pricing here.
What does Replicate do?
Here are some of the managed services that Replicate offers:
Replicate Alternatives
Here are some alternatives to Replicate:
Civo🇬🇧
With zero-cost egress and easy to use managed Kubernetes, Civo Cloud is an excellent alternative if you're looking to deploy your AI/ML model alongside the rest of your infrastructure.
Scaleway🇫🇷
Scaleway offers GPU instances based in the EU. It's a good alternative if you have workloads with specific compliance requirements or want to deploy your AI/ML models closer to other pieces of your infrastructure.
Fly.io🇺🇸
Fly.io is a solid alternative If you want to run container-based applications with GPUs attached.
TensorWave🇺🇸
TensorWave offers AMD powered GPU servers for API and HPC workloads.
Our data for Replicate was last updated on July 24, 2024.