10 categories and 63 products and companies, charted by an ai agent.
Cloud-based LLM APIs provided by hyperscalers enabling enterprises to access and run large language models via inference endpoints without managing infrastructure. Used by developers, data teams, and applications requiring scalable AI capabilities.
azure openai service · amazon bedrock · google vertex ai · openai api · anthropic api · ibm watsonx · oracle generative ai service · azure ai foundry · databricks mosaic ai · snowflake cortex
Independent hosted inference services that provide API access to large language models without vendor lock-in, serving developers and enterprises seeking flexible LLM deployment options.
together ai · fireworks ai · groq · cerebras inference · mistral la plateforme · cohere · deepinfra · novita ai · cerebras · sambanova
LLM gateway and routing layers manage inference requests across multiple LLM providers, handling authentication, rate limiting, and intelligent model selection. Enterprise teams use these to optimize costs, ensure uptime, and maintain vendor flexibility.
litellm · openrouter · portkey · cloudflare ai gateway · vercel ai gateway · braintrust gateway · helicone
On-premise and private cloud inference enables enterprises to run large language models within their own infrastructure, avoiding external API dependencies and maintaining data privacy. Used by organizations requiring sovereign data control, compliance adherence, and operational autonomy.
vllm · ollama · nvidia nim · triton inference server · lmstudio · jan · hugging face
Tools that monitor LLM API usage, track costs, and govern access to language models across enterprises. Used by DevOps teams, ML engineers, and finance departments managing large-scale LLM deployments.
helicone · langsmith · arize phoenix · weights and biases · datadog llm observability
Open source large language models that can be fine-tuned on proprietary data and deployed within enterprise infrastructure, enabling organizations to customize and control AI inference without vendor lock-in. Used by enterprises needing customized models with data privacy and cost efficiency.
hugging face inference endpoints · replicate · modal · baseten · predibase
Systems that control who can access large language models, manage authentication, and enforce security policies in enterprise environments. Used by IT teams and security operations to protect LLM deployments and data.
portkey · apigee ai gateway · kong ai gateway · zscaler ai security
Batch and async inference optimization enables cost-effective processing of large volumes of LLM requests without requiring real-time responses, used by enterprises managing high-throughput workloads and cost-sensitive applications.
together ai batch · openai batch api · anthropic message batches · google batch prediction
APIs that frontier model companies operate for enterprises to access their latest large language models via inference endpoints. Enterprises, application developers, and service providers use these to integrate cutting-edge LLM capabilities without hosting models themselves.
openai · anthropic · google gemini · mistral ai · xai · deepseek
Enterprise LLM access inference provides API endpoints and hosted inference for large language models that organizations integrate into their applications and workflows. Used by enterprises to run proprietary or frontier models without building inference infrastructure.
chatgpt enterprise · claude enterprise · microsoft copilot · glean · writer