Open-weight large language models (LLMs) are usually named as model artifacts, but production users often consume them as hosted API services. This paper argues that the operational unit is a service object: a provider-specific, time-varying endpoint defined by model variant, protocol behavior, context capacity, listed price, latency and throughput distribution, reliability, and task feasibility. Using sampled request logs, provider metadata, compatibility probes, pricing snapshots, and continuous latency measurements collected by AI Ping during Q4 2025, we study how this service layer changes the meaning of "the same model." Three empirical patterns emerge. First, observed demand is concentrated but persistent across versions: in the displayed family aggregate, the largest family carries 32.0% of relative demand and the top five carry 87.4%, with a Gini coefficient of 0.693, while older variants remain active after newer releases. Second, supply and use separate: provider listing breadth does not imply realized adoption, and listed prices are more anchored than latency, throughput, context length, protocol support, and error semantics. Third, task mix matters: applications induce different token-length regimes, so provider choice is a constrained decision over provider-model-task-time tuples rather than a lookup by model name. In two representative counterfactuals under observed feasibility constraints, routing lowers Qwen3-32B cost by 37.8% and raises DeepSeek-V3.2 average throughput by about 90% relative to direct official access. The results support a measurement view of hosted open-weight LLMs as heterogeneous services, not static catalog entries. We open-source the measurement methodology and reproduction artifacts at https://github.com/haoruilee/llm_api_measurement_study to support result reproduction.

翻译：开放权重的大语言模型通常以模型制品命名，但生产环境用户往往通过托管API服务消费它们。本文提出，操作单元应为服务对象：一个由提供者定义、随时间变化的端点，其属性包括模型变体、协议行为、上下文容量、标价、延迟与吞吐量分布、可靠性以及任务可行性。基于AI Ping在2025年第四季度收集的采样请求日志、提供者元数据、兼容性探测数据、定价快照以及持续延迟测量结果，我们研究了服务层如何改变"同一个模型"的含义。三项实证规律浮现。首先，观察到的需求集中但跨版本持续：在展示的家族聚合中，最大家族承载32.0%的相对需求，前五大家族承载87.4%，基尼系数为0.693，而旧变体在新版本发布后仍保持活跃。其次，供给与使用分离：提供者的列表广度不意味着实际采用，且标价比延迟、吞吐量、上下文长度、协议支持和错误语义更具锚定效应。第三，任务组合重要：应用诱发不同的令牌长度模式，因此提供者选择是对提供者-模型-任务-时间元组的约束决策，而非根据模型名称查找。在观测可行性约束下的两个代表性反事实场景中，相对于直接官方访问，路由使Qwen3-32B成本降低37.8%，使DeepSeek-V3.2平均吞吐量提升约90%。研究结果支持将托管开放权重LLM视为异质服务而非静态目录项的测量视角。我们在https://github.com/haoruilee/llm_api_measurement_study开源了测量方法和复现制品，以支持结果复现。