unslothai/unsloth · error · HTTPException

File too large ({size_bytes} bytes). Maximum is {UNSTRUCTURE

Error message

File too large ({size_bytes} bytes). Maximum is {UNSTRUCTURED_RECIPE_UPLOAD_MAX_LABEL}.

What it means

HTTP 413 raised by POST /seed/upload-unstructured-file when a single file exceeds UNSTRUCTURED_RECIPE_UPLOAD_MAX_BYTES. The whole body is read into memory first, then the per-file cap is enforced; the message reports the actual size and the human-readable max (UNSTRUCTURED_RECIPE_UPLOAD_MAX_LABEL).

Source

Thrown at studio/backend/routes/data_recipe/seed.py:457

) -> UnstructuredFileUploadResponse:
    _validate_safe_id(block_id, "block_id")

    original_filename = file.filename or "upload"
    ext = Path(original_filename).suffix.lower()
    if ext not in UNSTRUCTURED_ALLOWED_EXTS:
        raise HTTPException(
            400,
            f"Unsupported file type: {ext}. Allowed: {', '.join(sorted(UNSTRUCTURED_ALLOWED_EXTS))}",
        )

    content = await file.read()
    size_bytes = len(content)

    if size_bytes == 0:
        raise HTTPException(400, "Empty file not allowed")

    if size_bytes > UNSTRUCTURED_RECIPE_UPLOAD_MAX_BYTES:
        raise HTTPException(
            413,
            f"File too large ({size_bytes} bytes). Maximum is {UNSTRUCTURED_RECIPE_UPLOAD_MAX_LABEL}.",
        )

    block_dir = UNSTRUCTURED_UPLOAD_ROOT / block_id
    ensure_dir(block_dir)
    current_total = _get_block_total_size(block_dir)
    if current_total + size_bytes > UNSTRUCTURED_RECIPE_UPLOAD_TOTAL_MAX_BYTES:
        raise HTTPException(
            413,
            f"Total upload limit ({UNSTRUCTURED_RECIPE_UPLOAD_TOTAL_MAX_LABEL}) exceeded",
        )

    file_id = uuid4().hex
    raw_path = block_dir / f"{file_id}{ext}"
    raw_path.write_bytes(content)

    extracted_path = block_dir / f"{file_id}.extracted.txt"

View on GitHub (pinned to 203007d190)

Solutions

1. Split the document (extract chapters/ranges) and upload parts under the limit.
2. Reduce PDF size (compress images, e.g. gs or qpdf) before upload.
3. If policy allows, raise UNSTRUCTURED_RECIPE_UPLOAD_MAX_BYTES in backend config.

Example fix

# before
upload(myfile.pdf)  # 250MB, limit 100MB

# after
qpdf --split-pages myfile.pdf --out-file part  # then upload part-01.pdf etc.

const MAX = UNSTRUCTURED_MAX_BYTES; // keep in sync with backend config
if (file.size > MAX) throw new Error(`split ${file.name} (${file.size} > ${MAX})`);

function isUnderPerFileLimit(f: File, max: number): boolean { return f.size <= max; }

On 413 with the per-file message, prompt the user to split/compress that specific file; upload the remaining files that passed validation.

When it happens

Trigger: Uploading one large PDF/docx above the configured per-file limit (message states size and max explicitly).

Common situations: Scanned-book PDFs or large exports; environment variable tuning the limit down after big files were already in use; users unaware of the cap.

Related errors

• Total upload limit ({UNSTRUCTURED_RECIPE_UPLOAD_TOTAL_MAX_LA
• Dataset upload too large. Maximum is {get_upload_limit_label
• Unsupported file type: {ext}. Allowed: {', '.join(sorted(UNS
• File exceeds the {cap // (1024 * 1024)} MB upload limit.
• Image is too large ({w}x{h}); maximum is {max_side}px per si

AI-assisted analysis of unslothai/unsloth@203007d190 (2026-08-15). Data as JSON: /api/errors/371055b20b606411. Report an issue: GitHub.