Build faster your
Use open source LLMs or GenerativeAI models
Fine-tune on demand on your data
Benefit from HPC and serverless GPUs

RYAX is the new low-code serverless and open-source solution to build faster your AI workflows and applications

ryax-front-illu-befunky5

Serverless for Generative AI

Easy deployment of any script, code, container or AI model into production through serverless technologies, for CPUs and GPUs. Define CPU/GPU/RAM resources needed for each function

devops-pipeline

Low-code to build your AI application

Built-in reusable triggers & actions to get you started fast. Build & deploy any piece of custom code.
Speed up with AI code generation

consulting-data

Foundation Models

Fine-tuning of any open source Large Language Model or Generative Image Model with your private data, and inference of your model on GPU through Ryax APIs and workflows
Models available: StabilityAI/StableLM, Databricks/Dolly, MetaAI/SAM, HuggingFaceHub

chip

Execute on HPC

From Cloud/Kubernetes to HPC/Slurm, and back, execute each step of your workflows where you want

HPC

Workflows and APIs

Efficiently automate your AI application with workflows: on-event triggering, data loading, data pre-processing / post processing,  APIs requests, etc.

workflow ML-octopus

They trust us

sanofi
orange
ATOS2
keolis
UCIT
n'cloud
sogilis
INRIA2