Modal

Inference

Deploy and scale inference for LLMs, audio, image, and video generation workloads.

Training

Fine-tune open-source models on single or multi-node clusters instantly.

Sandboxes

Programmatically scale secure, ephemeral environments for running untrusted code.

Batch processing

Scale to thousands of containers for on-demand batch workloads.

Notebooks

Collaborate on code and data in real-time with shareable notebooks.

Audio transcription

Transcribe speech in batches using Whisper, turning audio bytes into text at scale.

Voice chat with LLMs

Build interactive voice chat applications.

Image and video inference

Run computational biology, image, and video inference tasks.

Music generation

Turn prompts into music with ACE-Step.

Text-to-speech

Deploy a TTS API with Chatterbox to generate natural audio from text.

Programmable infrastructure

Define everything in code—no YAML or config files—keeping environment and hardware requirements in sync.

Elastic GPU scaling

Access thousands of GPUs across clouds with no quotas or reservations, scaling back to zero when idle.

Unified observability

Integrated logging and full visibility into every function, container, and workload.

AI-native runtime

Engineered from the ground up for heavy AI workloads, with super-fast autoscaling and model initialization, claimed to be 100x faster than Docker.

Built-in storage layer

A globally distributed storage system built for high throughput and low latency, designed for fast model loading, training data, or other datasets.

First-party integrations

Mount existing cloud buckets, connect to MLOps tools, and send data to existing telemetry vendors.

Multi-cloud capacity pool

Deep multi-cloud capacity with intelligent scheduling ensures you always have the CPUs and GPUs you need without managing input orchestration.

Security and governance

Team controls, battle-tested isolation, SOC2 & HIPAA compliance, and data residency controls.