open-webui/open-webui · error · HTTPException

Mistral API key is required for Mistral TTS

Error message

Mistral API key is required for Mistral TTS

What it means

HTTP 400 raised in _tts_mistral (backend/open_webui/routers/audio.py) when the audio.tts.mistral.api_key config is empty. The Mistral TTS handler requires a Bearer token for POST {api_base_url}/audio/speech and fails fast before making the request.

Source

Thrown at backend/open_webui/routers/audio.py:513

            },
        )
        sf.write(str(file_path), wav['audio'], samplerate=wav['sampling_rate'])

    await asyncio.to_thread(_run_pipeline)

    # Audio file already written by sf.write; just persist the request metadata.
    async with aiofiles.open(file_body_path, 'w') as f:
        await f.write(json.dumps(payload))
    return FileResponse(file_path)


async def _tts_mistral(request, payload, file_path, file_body_path, user):
    """Generate speech via the Mistral TTS API."""
    api_key = await Config.get('audio.tts.mistral.api_key')
    api_base_url = await Config.get('audio.tts.mistral.api_base_url') or 'https://api.mistral.ai/v1'

    if not api_key:
        raise HTTPException(status_code=400, detail='Mistral API key is required for Mistral TTS')

    r = None
    try:
        session = await get_session()
        r = await session.post(
            url=f'{api_base_url}/audio/speech',
            json={
                'input': payload.get('input', ''),  # text to synthesize
                'model': await Config.get('audio.tts.model') or 'voxtral-mini-tts-2603',
                'voice_id': payload.get('voice', ''),
                'response_format': 'mp3',
            },
            headers={
                'Content-Type': 'application/json',
                'Authorization': f'Bearer {api_key}',
            },
            ssl=AIOHTTP_CLIENT_SESSION_SSL,
        )

View on GitHub (pinned to 01f4282f1f)

Solutions

1. Set the Mistral API key under Admin Settings > Audio (audio.tts.mistral.api_key) and retry.
2. If configured via environment/config file, ensure the key value is non-empty and the backend is restarted.
3. Confirm the key is valid by calling the Mistral API directly (e.g. their models endpoint) to rule out an auth mix-up.

Example fix

# before (env)
AUDIO_TTS_ENGINE=mistral
# no api key set

# after
AUDIO_TTS_ENGINE=mistral
AUDIO_TTS_MISTRAL_API_KEY=<your-key>

api_key = await Config.get('audio.tts.mistral.api_key')
if engine == 'mistral' and not api_key:
    raise HTTPException(400, 'Configure audio.tts.mistral.api_key before using Mistral TTS')

When it happens

Trigger: Selecting 'mistral' as audio.tts.engine in Admin Settings without entering a Mistral API key, or having the key stored as an empty string.

Common situations: Admin switched the TTS engine to Mistral but skipped the key field; the key was cleared during a config migration or reset; env-var-based key not passed into the container.

Related errors

• API key cannot be empty.
• No audio_data in Mistral TTS response
• We could not find what you're looking for :/
• Unsupported TTS engine: {engine}
• Mistral API key is required for Mistral STT

AI-assisted analysis of open-webui/open-webui@01f4282f1f (2026-08-14). Data as JSON: /api/errors/2d4d3774d7d5e80c. Report an issue: GitHub.