Zibra Labs

Description

Builds distributed CPU and GPU compute clusters for AI workloads, supporting massively parallel simulation, backtesting, parameter sweeps, post-training, and reinforcement learning.

Related

• Equal1Builds rack-mounted quantum computers that deploy alongside existing CPU and GPU compute; intended for data centers and teams operating…
• Standard MachinesBuilds environments and evaluations for advanced chip design, including reinforcement learning environments for long-horizon design tasks…
• EdotEnvBuilds scalable reinforcement learning settings using financial market data, helping AI labs and academic groups assess and post-train…
• LambdaBuilds modular AI factories and provides managed GPU clusters, supercomputers, and instances for labs and teams running AI training,…
• EnactBuilds a post-training layer for physical AI, generating robotics data, identifying policy failure states, creating targeted recovery…
• Together AIBuilds a full-stack AI platform offering inference APIs, batch inference, GPU clusters, custom infrastructure, training, fine-tuning, and…
• Voltage ParkProvides flexible GPU cloud compute, including on-demand and reserved access, for startups, enterprises, and research institutions building…
• TorusProvides engineering intelligence for capital projects by extracting parameters from documents, cross-referencing specifications, drawings,…
• Foundry CloudCloud platform for accessing GPU clusters and orchestrating AI workloads. Palo Alto CA HQ.
• Nodus ComputeBuilds an AI compute platform that places and runs training, inference, and agent workloads, with checkpoint recovery and an…
• TokenlessFinds flaws in reinforcement learning environments and evaluations, helping teams training models or running agent evals prevent reward…
• MagicBuilds large code models that aim to automate software engineering, AI research, and code generation using pretraining, reinforcement…