BoundaryML/baml · error

{e}

Error message

{e}

What it means

While orchestrating the stream, BAML renders the prompt for the current graph node; if rendering fails (missing/invalid params, context errors, template failures), the node yields LLMResponse::InternalFailure and the error string is propagated as-is. The message is the underlying render error, not an HTTP problem.

Source

Thrown at engine/baml-runtime/src/internal/llm_client/orchestrator/stream.rs:215

                results.push((
                    cancel_scope,
                    LLMResponse::Cancelled("Operation cancelled".to_string()),
                    Some(Err(anyhow::anyhow!(
                        crate::errors::ExposedError::AbortError {
                            detailed_message: String::new()
                        }
                    ))),
                ));
                break;
            }
            result = async {
                let prompt = match node.render_prompt(ir, prompt, ctx, params).await {
                    Ok(p) => p,
                    Err(e) => {
                        return Some((
                            node.scope,
                            LLMResponse::InternalFailure(e.to_string()),
                            Some(Err(anyhow::anyhow!(e.to_string()))),
                        ));
                    }
                };


                let (system_start, instant_start) = (web_time::SystemTime::now(), web_time::Instant::now());
                let ctx = CtxWithHttpRequestId::from(ctx);
                let stream_res = node.stream(&ctx, &prompt).await;
                let final_response = match stream_res {
                    Ok(mut response_stream) => {

                        let mut last_response: Option<LLMResponse> = None;
                        let parse_state = Arc::new(Mutex::new(ParserState::default()));
                        let (snapshot_tx, snapshot_rx) = watch::channel::<Option<Arc<LLMCompleteResponse>>>(None);

                        let parser_future = on_event.as_ref().map(|on_event_cb| {
                            let scope = node.scope.clone();
                            let parse_state = parse_state.clone();

View on GitHub (pinned to bd85ce9dee)

Solutions

1. Read the wrapped message {e} — it names the actual render failure
2. Ensure all required function parameters are supplied and correctly typed in the client call
3. Verify any ctx.* variables used in the prompt exist in the RuntimeContext passed in
4. Reproduce locally with a minimal call and print the params before invoking

Example fix

// before
await b.functions.Classify(); // missing required input
// after
await b.functions.Classify("some text", { ctx: { tenant: "acme" } });

function validateArgs(fnName, args, schema) {
  for (const [k, v] of Object.entries(schema)) {
    if (args[k] === undefined) throw new Error(`${fnName}: missing required param ${k}`);
    if (v.type === 'string' && typeof args[k] !== 'string') throw new Error(`${fnName}: param ${k} must be a string`);
  }
}
validateArgs('Extract', args, { text: { type: 'string' } });

const hasRequired = (args, keys) => keys.every(k => args[k] !== undefined && args[k] !== null);

When it happens

Trigger: A prompt node in the retry/orchestration graph fails render_prompt due to missing required function parameters, invalid context variables, or unresolvable Jinja-style expressions in the BAML prompt.

Common situations: Calling a BAML function without a required argument; passing a wrong-typed value (e.g. object where image expected); ctx variables referenced in the prompt not provided in the runtime context.

Understand the failure class

Background: "missing required argument" and "the following required arguments were not provided": what required-argument errors mean and how to fix them — this error's family across 20 libraries.

Related errors

• No model type supported
• no-color diagnostics do not invoke the message highlighter
• rendering without a message highlighter cannot fail semantic
• colored diagnostics have a source highlighter
• ai.Prompt._data must contain baml_builtins2::PromptAst

AI-assisted analysis of BoundaryML/baml@bd85ce9dee (2026-09-12). Data as JSON: /api/errors/b9f30e8b6618bef1. Report an issue: GitHub.