NVDANVIDIA Corporation

Appears in these Strategy Maps

• Databricks

  SupplierPrice could fallInferred14 conn

  GPU infrastructure dependency for Databricks ML runtime and AI workload performance, where NVIDIA accelerators accessed through cloud provider GPU instances.

  Evidence as of 2026-07-17

• Anysphere

  SupplierPrice could fallInferred8 conn

  GPU infrastructure powering real-time LLM inference for code suggestions at scale; NVIDIA dominates AI inference hardware market that Anysphere's cloud providers rely on.

  Evidence as of 2026-07-18

• xAI

  SupplierPrice could riseInferred18 conn

  Direct supplier of tens of thousands of H100/H200 GPUs powering xAI's Colossus supercomputer and all Grok model training workloads.

  Evidence as of 2026-07-17

• Shield AI

  Infrastructure providerPrice could riseInferred6 conn

  NVIDIA GPU dependence for Hivemind AI training and inference; demand for NVIDIA data center and edge AI hardware driven by defense AI adoption

  Evidence as of 2026-07-17

• Harvey

  Infrastructure providerPrice could fallInferred10 conn

  Harvey's dependence on NVIDIA GPU availability (H100/A100 chips) for LLM inference and fine-tuning through cloud providers.

  Evidence as of 2026-07-20

• Neuralink

  SupplierPrice could riseEvidence aging5 conn

  AI compute infrastructure powering neural decoding algorithms and machine learning models that process high-bandwidth electrode signals; GPU architecture optimized for parallel signal processing workloads.

  Evidence as of 2023-11-08

• ByteDance

  SupplierPrice could riseEvidence aging19 conn

  Critical supplier of AI training and inference GPUs essential for ByteDance's recommendation algorithms and content processing at billion-user scale.

  Evidence as of 2023-10-17

• Mistral AI

  SupplierPrice could riseInferred22 conn

  Direct upstream dependency on NVIDIA H100/A100 GPUs essential for all model training and inference operations

  Evidence as of 2026-07-20

• Cohere

  SupplierPrice could fallInferred23 conn

  Cohere's critical dependency on NVIDIA H100/A100 GPUs for training and inference of large language models at scale.

  Evidence as of 2026-07-21

• TerraPower

  Infrastructure providerPrice could riseLikely7 conn

  Technology partnership exposure via March 25, 2026 integration of NVIDIA Omniverse platform (with SoftServe) to accelerate TerraPower's reactor design, digital twin modeling, and deployment workflows; NVIDIA's simulation and AI tools embedded in TerraPower's engineering stack.

  Evidence as of 2026-03-26

• Epic Games

  Infrastructure providerPrice could riseEvidence aging9 conn

  Unreal Engine development and optimization relies heavily on Nvidia GPU architectures (RTX ray tracing, DLSS), plus Epic uses Nvidia data center GPUs for rendering and AI workloads.

  Evidence as of 2023-03-21

• Valve Corporation

  SupplierPrice could fallInferred11 conn

  Direct silicon and GPU partnership for Steam Deck APU design collaboration with AMD, and critical graphics driver integration for Steam platform and VR Index headset; Nvidia GPUs power majority of high-end Steam gaming PCs and VR systems

  Evidence as of 2026-07-20

• ElevenLabs

  Infrastructure providerPrice could fallInferred21 conn

  NVIDIA GPU availability and pricing directly determines ElevenLabs' training costs, inference latency, and capacity to scale voice synthesis model quality and customer throughput.

  Evidence as of 2026-07-21

• Cohere

  SupplierPrice could fallInferred23 conn

  Cohere's critical dependency on NVIDIA H100/A100 GPUs for training and inference of large language models at scale.

  Evidence as of 2026-07-21

• Telegram

  SupplierPrice could riseInferred9 conn

  Telegram's rapid AI feature expansion (AI bots, AI summaries, AI text generation per June-July 2026 updates) requires GPU compute infrastructure; NVIDIA supplies data center GPUs (H100, A100) that power inference workloads for Telegram's AI Editor, bot-to-bot chats, and 100M+ AI-driven emoji/sticker search functionality.

  Evidence as of 2026-07-14

• Runway

  SupplierPrice could fallInferred12 conn

  Runway's generative video models depend entirely on NVIDIA H100/A100 GPUs for both training and real-time inference at scale.

  Evidence as of 2026-07-23

• Safe Superintelligence

  SupplierPrice could fallEvidence aging20 conn

  NVIDIA H100/H200 GPUs are sole viable hardware for training superintelligent models at SSI's required scale; direct dependency documented in definition.

  Evidence as of 2024-03-01

• Helsing

  SupplierDirection unclearEvidence aging16 conn

  Critical supplier of high-performance AI GPUs required for training and deploying defense ML models; Helsing's compute infrastructure costs and performance directly tied to Nvidia accelerator availability and pricing.

  Evidence as of 2024-11-20