_on_llm_start() — langchain Function Reference

Architecture documentation for the _on_llm_start() function in stdout.py from the langchain codebase.

Function python Observability Tracers calls 2

Entity Profile

Observability→ Tracers→ _on_llm_start() — langchain Function Reference

Dependency Diagram

graph TD
  3847ea36_9090_74f7_44f0_b0626355ef3b["_on_llm_start()"]
  56d9fac7_db71_9c71_7cac_1d680d392fa0["FunctionCallbackHandler"]
  3847ea36_9090_74f7_44f0_b0626355ef3b -->|defined in| 56d9fac7_db71_9c71_7cac_1d680d392fa0
  1bda6774_d332_dec5_ed8f_8d6eef57961e["get_breadcrumbs()"]
  3847ea36_9090_74f7_44f0_b0626355ef3b -->|calls| 1bda6774_d332_dec5_ed8f_8d6eef57961e
  95875f96_0c2d_b364_e63a_651f8ed4cdcd["try_json_stringify()"]
  3847ea36_9090_74f7_44f0_b0626355ef3b -->|calls| 95875f96_0c2d_b364_e63a_651f8ed4cdcd
  style 3847ea36_9090_74f7_44f0_b0626355ef3b fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

libs/core/langchain_core/tracers/stdout.py lines 136–147

    def _on_llm_start(self, run: Run) -> None:
        crumbs = self.get_breadcrumbs(run)
        inputs = (
            {"prompts": [p.strip() for p in run.inputs["prompts"]]}
            if "prompts" in run.inputs
            else run.inputs
        )
        self.function_callback(
            f"{get_colored_text('[llm/start]', color='green')} "
            + get_bolded_text(f"[{crumbs}] Entering LLM run with input:\n")
            + f"{try_json_stringify(inputs, '[inputs]')}"
        )