Note Do you want to add yourself to this list? Simply fork the repository and open a PR with the required change.
| Type | Name | Since | Website | Description |
|---|---|---|---|---|
| Founding Contributor | CoreWeave | 2025 | CoreWeave | CoreWeave is a founding contributor to llm-d, building on their long-standing commitment to open source AI and scalable inference infrastructure. |
| Founding Contributor | Google Cloud | 2025 | Google Cloud | Google Cloud is a founding contributor to llm-d, contributing to distributed AI inference at scale and support for Google Cloud TPUs in vLLM. |
| Founding Contributor | IBM Research | 2025 | IBM Research | IBM Research is a founding contributor to llm-d, focused on building a hardware-agnostic distributed AI inference platform for enterprise workloads. |
| Founding Contributor | NVIDIA | 2025 | NVIDIA | NVIDIA is a founding contributor to llm-d, accelerating the project with innovations from NVIDIA Dynamo including the NVIDIA Inference Xfer Library (NIXL). |
| Founding Contributor | Red Hat | 2025 | Red Hat | Red Hat is a founding contributor to llm-d, providing project infrastructure and community stewardship in collaboration with the broader contributor community. |
| Launch Partner | AMD | 2025 | AMD | AMD contributes high-performance GPU expertise to advance AI inference for evolving enterprise AI needs. |
| Launch Partner | Cisco | 2025 | Cisco | Cisco is exploring real-world use cases that help organizations apply AI more effectively and efficiently through llm-d. |
| Launch Partner | Hugging Face | 2025 | Hugging Face | Hugging Face supports llm-d to help developers take AI applications built on the Hugging Face ecosystem to scale. |
| Launch Partner | Intel | 2025 | Intel | Intel is a launch partner supporting llm-d as the latest milestone in a decades-long collaboration with Red Hat to empower enterprises with open source solutions. |
| Launch Partner | Lambda | 2025 | Lambda | Lambda supports llm-d to make state-of-the-art inference accessible, efficient, and open. |
| Launch Partner | Mistral AI | 2025 | Mistral AI | Mistral AI is a launch partner supporting the llm-d community for distributed generative AI inference at scale. |
| Contributor | DaoCloud | 2025 | DaoCloud | DaoCloud contributes to llm-d, leveraging P/D disaggregation and KV-cache architectures in its d.run MaaS platform. |
| Contributor | Moreh | 2025 | Moreh | Moreh contributes to llm-d, actively improving KV-cache management and vLLM rendering, along with P/D disaggregation for distributed inference. |
| Contributor | Rebellions | 2025 | Rebellions | Rebellions contributes to llm-d, enabling support for the NPU accelerator to bring high-performance, energy-efficient AI inference to the platform. |
| University Supporter | University of California, Berkeley (Sky Computing Lab) | 2025 | Sky Computing Lab | The Sky Computing Lab at UC Berkeley, originators of vLLM, are founding academic supporters of llm-d. |
| University Supporter | University of Chicago (LMCache Lab) | 2025 | LMCache Lab | The LMCache Lab at the University of Chicago, originators of LMCache, are founding academic supporters of llm-d. |
| User | Tesla | 2025 | Tesla | The Tesla ML Platform team are users of llm-d. |
| User | Cohere | 2026 | Cohere | The Cohere model serving and RL team are users of llm-d. |
| User | JetBrains | 2026 | JetBrains | The JetBrains model serving team are users of llm-d. |
| User | Stelia | 2026 | Stelia.ai | Stelia.ai's managed inference service is powered by llm-d. |
| User | Reflection | 2026 | Reflection.ai | Reflection uses llm-d's async processing for our internal inference platform. |
| User | Prime Intellect | 2026 | Prime Intellect | Prime Intellect uses llm-d in prime-rl for multi-node WideEP RL rollouts. |
| User | Snowflake | 2026 | Snowflake | The Snowflake Cortex team are users of llm-d. |
| User | Novita AI | 2026 | Novita | The Novita AI team are users of llm-d. |
| User / Contributor | DigitalOcean | 2026 | DigitalOcean | DigitalOcean uses llm-d for Dedicated and Serverless Inference endpoints and contributes actively to the project. |
| User | Capital One | 2026 | Capital One | The Capital One model serving team are users of llm-d. |
Founding Contributor: The organization is a founding contributor to the llm-d project, actively contributing from the project's inception.
Launch Partner: The organization is a supporting partner at the launch of the llm-d project, contributing expertise and collaboration to the community.
Contributor: The organization actively contributes code, documentation, or community resources to the llm-d project.
University Supporter: An academic institution supporting the llm-d project, often through originating or contributing to foundational open source technologies used by llm-d.
User: The organization is a user of the llm-d project, contributing feedback and real-world examples to the community.