A research report released by Huatai Securities highlights a notable transition in the competitive landscape of foundational models. This shift is moving away from the initial phase of product exploration towards a phase of commercialization, propelled by high-value scenarios, the enterprise market, and substantial usage volume. In the nascent stages of this domain, OpenAI successfully garnered a vast consumer base by leveraging bottom-up product innovation through ChatGPT. Concurrently, Anthropic dedicated its efforts to bolstering model capabilities and crafting Coding and enterprise APIs, thereby gradually cementing its advantages in resource allocation as the Coding scenario reached maturity. Meanwhile, firms such as Zhipu and MiniMax are also progressively honing in on the API, Coding, and Agent sectors. Huatai Securities posits that the competition within the foundational model industry has transcended mere model capabilities and user base size. Instead, it has broadened to encompass a multifaceted array of dimensions, including the consumption of high-value tokens, the depth of enterprise market penetration, unit economics, and the efficacy of product distribution strategies.