Runtime integration hub
Private AI Runtimes
Choose production inference, lab endpoints, internal utility models, and media workflows according to workload and operating responsibility.
Key considerations
vLLM
GPU-backed production chat inference.
LM Studio
Local labs and smaller evaluation environments.
Ollama and ComfyUI
Internal utility work and governed media workflows.
The workspace and runtime are separate responsibilities
Pulsar provides the user, policy, workflow, storage, and operating layer. Model and media services can be selected, sized, and managed independently.
Use the decision matrix
Choose a runtime according to workload, hardware, concurrency, latency, model governance, and the team responsible for keeping it healthy.
Pulsar with ComfyUI
Connect ComfyUI for governed image creation, editing, video workflows, model assets, queue state, and retained outputs.
Read morePulsar with LM Studio
Connect an LM Studio OpenAI-compatible endpoint for local labs, evaluation environments, smaller deployments, and model testing.
Read morePulsar with Ollama
Use an optional CPU-friendly Ollama service for internal titles, summaries, prompt optimization, scheduler summaries, and diagnostics.
Read morePulsar with vLLM
Use vLLM as a private OpenAI-compatible model runner with backend-owned capacity control and operator-visible diagnostics.
Read morePrivate deployment consultation
Review Pulsar against your environment.
Bring the infrastructure, security boundaries, model runners, and use cases. The Pulsar team will map the appropriate deployment path.