Latest stories
Doppel improved its machine learning operations by migrating to Modal, achieving faster experimentation and more efficient inference with reduced infrastructure...
Runway integrates Modal's serverless computing to enhance real-time video capabilities for its Characters API, enabling diverse applications in various industri...
NVIDIA's RTX Pro 6000 Blackwell is now integrated into Modal, alongside updates to the Command Palette and Sandbox Filesystem API, enhancing user experience and...
Butter's acquisition by Modal brings new expertise in code generation and agent engineering, aiming to improve Modal Sandboxes' functionality.
Modal enhances AI research efficiency by integrating elastic GPU provisioning, enabling dynamic resource allocation for research workloads.
OpenAI's new Agents SDK enables developers to build tailored coding agents and internal tools efficiently, leveraging Modal's powerful sandbox environment.
This article discusses the implementation of a reinforcement learning workflow for automated theorem proving using the Modal platform, focusing on efficiency an...
SGLang v0.5.10 achieves over 16% throughput improvement and 10% latency reduction for multimodal inference through optimized cache lookups, enhancing efficiency...
Modal’s approach to serverless GPUs enhances allocation utilization and minimizes start times, addressing the challenges of dynamic inference workloads.
The collaboration between Anthropic and Modal introduces Claude Managed Agents, offering developers scalable, cost-effective AI workflows with integrated sandbo...
Applied Compute leverages reinforcement learning to develop tailored AI agents. This approach emphasizes adaptability, heavy reliance on proprietary data, and c...
New Role-Based Access Control (RBAC) features on Modal provide a secure framework for managing agent and human access to resources within environments.
This article examines the critical infrastructure challenges faced in reinforcement learning (RL) training and presents strategic solutions to optimize performa...
Optimizations to FlashAttention-4 target improved inference efficiency, with a focus on lower precision inputs and enhanced memory access strategies.
Modal introduces significant enhancements including regional routing for reduced latency, VM sandboxes for advanced workloads, and domain allowlisting for enhan...
Speculative decoding boosts AI model performance, delivering significant speed improvements and better handling of long-context tasks for practical applications...
Achieving faster sandbox user experiences requires addressing the entire startup lifecycle, from initial container booting to application readiness.
Modal's Auto Endpoints give teams the ability to manage and optimize their own inference setups, emphasizing cost-effective solutions without sacrificing perfor...
Modal’s new speculative decoding approach enhances inference speed, beating proprietary systems by over 60ms, showcasing the power of optimized AI inference.
Modal's new serverless server capabilities significantly enhance low-latency performance for applications, reducing p50 latency to 6ms while maintaining core fe...