agentscope-ai/agentscope · error · ValueError

Text embedding model {self.model!r} only accepts str inputs,

Error message

Text embedding model {self.model!r} only accepts str inputs, got {type(item).__name__}.

What it means

DashScope text embedding models accept only string inputs; _call_text type-checks every element of the batch and raises ValueError on the first non-str item, reporting its type name.

Source

Thrown at src/agentscope/embedding/_dashscope/_model.py:345

        self,
        inputs: list[str | DataBlock],
        **kwargs: Any,
    ) -> EmbeddingResponse:
        """Call the DashScope text embedding API for a single batch.

        Args:
            inputs (`list[str | DataBlock]`):
                Must all be ``str``; raises ``ValueError`` otherwise.
            **kwargs:
                Forwarded to the API.

        Returns:
            `EmbeddingResponse`: Embedding vectors and usage info.
        """
        texts: list[str] = []
        for item in inputs:
            if not isinstance(item, str):
                raise ValueError(
                    f"Text embedding model {self.model!r} only accepts "
                    f"str inputs, got {type(item).__name__}.",
                )
            texts.append(item)

        api_kwargs: dict[str, Any] = {
            "input": texts,
            "model": self.model,
            "dimension": self.dimensions,
            **kwargs,
        }

        if self.embedding_cache:
            cached = await self.embedding_cache.retrieve(
                identifier=api_kwargs,
            )
            if cached:
                return EmbeddingResponse(

View on GitHub (pinned to e90f1c7592)

Solutions

1. Convert all inputs to str before calling (decode bytes, stringify paths)
2. Filter or default None values to "" if your pipeline can emit them
3. Use a multimodal-capable model if you need to embed images/video

Example fix

# before
vecs = await model([b"hello", "world"])
# after
vecs = await model([x.decode("utf-8") if isinstance(x, bytes) else str(x) for x in items])

inputs = [x.decode() if isinstance(x, bytes) else x for x in inputs]

def all_str(items: list) -> bool:\n    return all(isinstance(i, str) for i in items)

try:\n    await model(texts)\nexcept ValueError as e:\n    if \"only accepts str\" in str(e): texts = [str(t) for t in texts]; await model(texts)\n    else: raise

When it happens

Trigger: Calling the model with a list containing non-str entries, e.g. bytes, DataBlock, dict, pathlib.Path, or None, via __call__ with text-mode inputs.

Common situations: Passing file contents read in binary mode (bytes); passing multimodal DataBlock objects to a text-only model; None values from upstream data pipelines with missing fields.

Related errors

• Invalid input: {item!r}. Expected str or DataBlock.
• DashScope text embedding API error: {response}
• DashScope multimodal embedding API error: {res}
• Multimodal embedding API only supports URL input for video d
• Unsupported media type {media_type!r} in DataBlock. Expected

AI-assisted analysis of agentscope-ai/agentscope@e90f1c7592 (2026-08-28). Data as JSON: /api/errors/1ed0b5bdd8432544. Report an issue: GitHub.