Unverified50% confidenceFactExact time
大模型技术演进分为大语言模型阶段、工具调用阶段和Agent阶段，第一阶段以GPT上线为标志，涌现OpenCompass、Evascope等评测工具及GSM8K数据集
1
Sources
50%
Confidence
Long-term
Relevance
7/25/2026
First Seen
Sources
Agent测试全解析：五大核心维度与自动化实战指南
Related Claims
UnverifiedThe GPT series follows the decoder route, adhering to the autoregressive language model objective, and has revealed the phenomenon of emergent capabilities through scaling.75% similarUnverifiedMainstream large models such as the GPT series, LLaMA, and DeepSeek are all built on the autoregressive language model paradigm.73% similarUnverifiedBy calling APIs of large models such as OpenAI's GPT series, Anthropic's Claude, and Google's Gemini, developers can build functional prototypes with natural language understanding, multi-turn dialogue, code generation, or image analysis capabilities within a few hours.70% similarUnverifiedGPT-OSS models are prepared for Agent scenarios and support tool-calling capabilities such as web search and code execution.70% similarUnverifiedOpenAI has chosen to develop reasoning models (o series) and conversational models (GPT series) separately.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/612289API
curl https://kongchang.com/api/v1/knowledge/claims/612289MCP
get_claim(id=612289)