{"list":"AI Infra 100","list_slug":"ai-infra","facet":"segment","value":"Inference & serving clouds","edition":"November 2026","published_at":"2026-11-01T13:18:02.499959+00:00","count":3,"of":100,"url":"https://www.frontiertechintelligence.com/leaders/ai-infra/segment/inference-serving-clouds","members":[{"rank":94,"name":"Baseten","url":"https://www.frontiertechintelligence.com/companies/baseten","category":"Compute software","segment":"Inference & serving clouds","country":"United States","score":80,"positioning":"Baseten provides hosted infrastructure for deploying and serving AI models, including open-weight large language models and NVIDIA-optimized inference workloads."},{"rank":95,"name":"Fireworks AI","url":"https://www.frontiertechintelligence.com/companies/fireworks-ai","category":"Compute software","segment":"Inference & serving clouds","country":"United States","score":80,"positioning":"Fireworks AI runs a hosted platform for deploying and serving foundation models such as Mistral and Kimi K2, selling inference infrastructure and APIs."},{"rank":96,"name":"OpenRouter","url":"https://www.frontiertechintelligence.com/companies/openrouter","category":"Compute software","segment":"Inference & serving clouds","country":"United States","score":76,"positioning":"OpenRouter operates a model gateway that provides API access to more than 400 AI models and processes over 10 trillion tokens per day. It monetizes this inference access through prepaid credits and usage charges."}]}