How Abundant designed efficient rollout architecture with Modal SandboxesAbundant builds reinforcement learning environments that frontier labs like OpenAI, Anthropic, Google, and Meta use to train and evaluate their models. Labs improve agent capability by giving them difficult tasks that Abundant designs, lettModal11/19 08:00
Building new research tools for Maya epigraphy on ModalMiguel Novelo grew up surrounded by Maya glyphs carved into stone across Mexico's Yucatán peninsula, printed in books, and painted on ceramics. For centuries, Maya scribes used this writing system to record histories, trace the movements ofModal11/18 08:00
Scaling Intelligence: A landmark commission and exhibitionLast week, Modal and Gray Area , the San Francisco nonprofit that has supported artists working at the intersection of art and technology for nearly two decades, opened Scaling Intelligence , an exhibition of 11 new artworks made with machiModal11/13 08:00
Sandbox resourcing: burst, don't reserveAcross Modal, a typical Sandbox spends more than half its life using less than 6% of its peak CPU. Take a coding agent: the Sandbox sits nearly idle while the model reasons about what to do next, and spikes only when the agent installs depeModal11/13 01:00
Modal becomes a gVisor maintainerThis week at KubeCon in Salt Lake City, Google is donating gVisor , including its name and trademarks, to the Cloud Native Computing Foundation (CNCF) , and the project is moving to a maintainer-led governance model. Modal is one of three cModal11/11 20:00
How Snorkel runs realistic agent simulations with ModalSnorkel , a frontier AI data lab, develops and validates complex datasets to train and evaluate AI systems. Snorkel’s Agentic Data Platform combines human experts with specialized, continually learning agents to produce higher quality data,Modal11/06 08:00
Launching a trillion-parameter model at Sakana AIDeploying a ~1-trillion parameter Mixture-of-Experts (MoE) model for API inference is a classic systems challenge. At full FP16 or BF16 precision, the memory footprint requires very large clusters. Standard post-training quantization (PTQ) Modal10/23 08:00
Introducing our Training APIThinking Machines’ Tinker changed how people do post-training. You write the training loop in plain Python on your laptop, and Tinker runs the forward and backward passes on its GPUs. We’re big fans, and so are many of the teams we work witModal10/22 20:00
Get started with LLM training on ModalWe’ve written about how reinforcement learning (RL) is an infrastructure problem , and how the last five years of work we’ve done provides performant , available , and scalable infrastructure you can rely on to train your models. Even withoModal10/21 20:00
Serverless routing made stickyAt Modal, our serverless compute powers real-time inference, training, and batch jobs, scaling automatically with demand. Sandboxes offer a different model, containers whose lifecycles you explicitly control. A growing number of workloads rModal10/20 08:00
NVIDIA Vera CPUs double throughput for Modal SandboxesLast week, NVIDIA hand-delivered Modal’s first Vera CPU server to our SF office. It’s now purring away in a server rack, and we’re pretty excited about what it could mean for Modal Sandboxes and, consequently, Modal users. Vera is NVIDIA’s Modal10/19 22:01
How Legora runs 2,000 parallel agent evals on Modal SandboxesLegora is an agentic system that supports lawyers in research, review, and drafting across complex matters. In early 2026, as rapid gains in model capabilities made longer and more autonomous legal work possible, Legora began replacing its Modal10/19 08:00
Building fast filesystems for VMsAt Modal, we like to think that we know a thing or two about squeezing nanoseconds out of systems. We love to obsess over making our systems performant and capable from the ground up, and we have to! We build for the thousands of GPUs waitiModal10/16 20:00
Augment Your LiveKit Agent with Modal GPUsLiveKit and their Agents SDK are a popular choice for implementing WebRTC-based audio-video applications. In this article, we share how you can use open-weights models or custom ones running on Modal alongside LiveKit agents. We’ll use BitHModal10/15 20:00
Muse Spark Open is now available on ModalMeta just released weights for Muse Spark Open, and it’s now available on Modal Auto Endpoints. Muse Spark Open is built for agentic coding and long-horizon tasks, with native multimodal understanding and a 1M-token context window. It suppoModal10/15 08:00
Modal crosses $1B in Annualized Run Rate and raises $750M Series DJust five months ago, when we announced our Series C, I wrote that demand was accelerating . Since then, usage has tripled and we’ve grown more than 15x year-over-year. We recently crossed $1B in annualized run rate with no sign of slowing Modal10/14 08:00
Toward RL on mixed-precision, mixed-accelerator samplersIn large-model reinforcement learning (RL), samplers usually run on the same accelerators and often the same precision as the trainer, keeping the two numerically close for stable training. Outside of RL, however, serving systems are not coModal10/13 20:00
Endpoint Candidates: evaluate and promote inference endpointsLatency, throughput, and cost optimization are at the core of modern inference engineering. Without testing and measurement, these metrics cannot be reliably improved. We’ve been focused on inference performance for some time now , and we’rModal10/12 08:00
TypeSafe AI scales Jev to trillions of tokens in three days on ModalTypeSafe AI ’s manifesto is contrarian for an AI lab: “Build Prod, Not God.” Their first model, Jev, delivers fast, structured decisions with frontier intelligence. In the course of a launch week, demand for Jev went stratospheric. Modal heModal10/05 08:00
Modal Clusters are generally availableOrganizations need to own their intelligence to be successful. Training and serving that intelligence at scale, however, requires petaFLOP/s of compute and terabit/s of networking, spread across many nodes. But owning intelligence doesn’t nModal10/01 20:00
VM Sandboxes: Full computers for agentsToday we’re making VM Sandboxes generally available on Modal, built for those who need to give their agents the power of a full computer. import modal app = modal.App.lookup("sandbox-manager", create_if_missing=True) sb = modal.Sandbox.creaModal10/01 08:00