Roadmap

See what we're working on and what's next

Available Now

Everything you need to start building today.

Compute Engine

Run containerized workloads in secure, isolated Firecracker microVMs.

Code Interpreter

Execute Python, JavaScript, and Bash code in sandboxed environments.

SDKs

Official libraries for Python and JavaScript/TypeScript.

REST API

Programmatic access to manage resources, accounts, and analytics.

Coming Soon

2025 / 2026

Exciting features and improvements on our roadmap.

GPU Compute

Coming Soon

On-demand NVIDIA A100, H100, and V100 GPUs for training, inference, and parallel processing.

Vector Databases

Coming Soon

High-performance vector storage and similarity search. Compatible with Pinecone, Weaviate, and more.

AI Storage

Coming Soon

Intelligent storage with auto-tiering, deduplication, and dataset versioning for ML workflows.

Model Registry

Coming Soon

Centralized model management with versioning, metadata tracking, and automated deployment.

Inference APIs

Coming Soon

Auto-scaling model serving with load balancing, A/B testing, and canary deployments.

Data Pipelines

Coming Soon

ETL/ELT workflows with real-time processing, data validation, and lineage tracking.

Agent Orchestration

Coming Soon

Multi-agent coordination with workflow management, resource sharing, and failure recovery.