# Gradium Voice Cloning and Voice Design (slim)

> Gradium's voice API creates a custom voice from an uploaded audio sample of about 10 seconds, or designs one from a written description, for its text-to-speech models in English, French, German, Spanish and Portuguese.

- Full: https://www.anchorterminal.com/tools/gradium-voice-cloning.md (~7,550 tokens) · this version ~1,780 tokens · JSON https://www.anchorterminal.com/tools/gradium-voice-cloning.json · canonical https://www.anchorterminal.com/tools/gradium-voice-cloning
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-07 · pricing https://www.anchorterminal.com/tools/gradium-voice-cloning#pricing · reviews https://www.anchorterminal.com/tools/gradium-voice-cloning#reviews
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

**C · 60.8/100 · rank #637 of 1486 · #6 in Voice cloning & custom voices · not agent-ready · confidence medium**


Assessment: A public OpenAPI file, dated release notes and a free plan with five custom voices and no card make Gradium's clone and design endpoints easy to adopt. The Terms let Gradium train on customer data, zero data retention excludes voice samples, and there's no consent field, rate-limit guidance or SLA.

## Facts

- Kind: Model API · vendor: Gradium SAS · category: Voice cloning & custom voices · legal entity: Gradium SAS · provenance 74/100
- Endpoint: `https://api.gradium.ai/api/voices/` (HTTP, websocket)
- Auth: API key · pricing: Freemium · x402: no · licence: Proprietary service under Gradium's Terms of Service (version of 19 October 2026). The `gradium` Python SDK is on PyPI
- Benchmark probes: not run yet, so Performance is pending
- Live, from our pollers (not part of the score): up at 2026-11-07 18:20 UTC · uptime 30 days 100.0% (200 probes) · p50 55 ms · p95 126 ms over 24 hours
- Vendor status page: no problems reported, All Systems Operational (2026-11-07 18:21 UTC)
- Surface listed: The voice endpoints of the Gradium API at api.gradium.ai (`POST /api/voices/` and the voice-generator routes), read from docs.gradium.ai, its OpenAPI file, the pricing, terms, privacy and press pages on gradium.ai and gradium.statuspage.io on 6 November 2026. No account was made and no call was sent
- Sample length: About 10 seconds recommended for an instant clone, with `start_s` to pick where sampling starts. Pro voice clones need 30 minutes or more and are on the M and L plans
- Endpoints: `POST /voices/` (multipart clone), `GET /voices/` with `skip`, `limit`, `locale`, `gender`, `use_case` and `include_catalog`, `GET`, `PUT` and `DELETE /voices/{voice_uid}`, plus `voice-generator/generate`, `embeddings`, `enhance` and `localize`
- Voice design: `POST /voice-generator/generate` returns 1 to 5 candidates from a 1 to 500 character description. Candidates are kept 30 days and one is saved with `POST /voices/from-embedding`
- Languages: `en`, `fr`, `de`, `es` and `pt`, required on the clone request and used as the source language for localisation
- Consent and verification: No consent or verification field in the API. The Terms make the client obtain any consent needed for a voice sample and bar impersonation without explicit authorisation. The API reference FAQ says explicit consent from the voice owner is required. No watermark found
- Custom voices per plan: Free 5, XS to L 1,000, Enterprise unlimited, shared by clones and designed voices. Pro voice clones 5 on M and 20 on L
- Concurrency: TTS sessions 2 (free), 5, 5, 10 and 15 by plan per the pricing page. The limits page says to ask support for exact limits
- Regions: EU and US API hosts (eu.api.gradium.ai, us.api.gradium.ai), with data residency from the XS plan
- Legal entity: Gradium, a société par actions simplifiée registered in Paris under 989 284 955, 151bis rue Saint-Honoré, 75001 Paris, per the Terms
- Prices: XS plan $12 per month (plan); S plan $39 per month (plan); M plan $312 per month (plan); L plan $1480 per month (plan)

- Scores: Reliability 63, Performance pending, Schema & documentation 83, Agent ergonomics 67, Security & auth 48, Payments & pricing 30, Task success pending, Maintenance & community 70, Transparency & trust 63 · total over the 7 assessed categories
- Why: Reliability, Graded on the hosted voice API with the hosted lines. · Schema & documentation, Public OpenAPI 3.1 file at docs.gradium.ai/api-reference/openapi.json (25). · Agent ergonomics, Voice objects are compact metadata (20 of 25). · Security & auth, Graded for voice cloning as the other listings in this category, with consent and misuse controls in place of the read-only line and trainin… · Payments & pricing, No x402, MPP or L402 (0). · Maintenance & community, Read as a model, on the lines the other hosted cloning listings use (recency 30, dated releases 10, deprecation notice 10, support 15, SDKs… · Transparency & trust, Closed service with clear terms, version of 19 October 2026, naming Gradium SAS with its Paris registration (15 of 30).
- Sources: 14, open questions: 6, both in the full twin
- Capabilities: voice.clone, voice.design, speech.tts
- JSON: https://www.anchorterminal.com/api/v1/tools/gradium-voice-cloning.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/gradium-voice-cloning.svg` or a link to https://www.anchorterminal.com/tools/gradium-voice-cloning from a page on gradium.ai or one of its subdomains, or the README at the root of github.com/gradium-ai/gradium-py, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.
- Support: Discord https://discord.gg/T85yt8kD33; support@gradium.ai; https://docs.gradium.ai/; https://gradium.ai/pricing-faq; more in the full twin

## Before you call it

1. Send `language` on every clone. It's required and must be one of en, fr, de, es or pt and match the recording
2. Clones are enhanced by default since 23 October 2026. Pass `enhance=false` to keep the raw voice, then check `enhance` on GET
3. Pass a clone as `voice_id` in TTS, not `voice`
4. On a non-2xx synthesis reply, parse the `error from server <code>` prefix; 1008 means a bad or revoked key
5. Check the custom-voice allowance first. The free plan holds 5 voices shared by clones and designs

## Connect

```bash
pip install gradium
```

```bash
curl -X POST https://api.gradium.ai/api/voices/ \
  -H "x-api-key: $GRADIUM_API_KEY" \
  -F "audio_file=@my_voice_sample.wav" \
  -F "name=My Custom Voice" \
  -F "language=en"
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| ElevenLabs Voice Cloning and Voice Design API | BB | 76.8 | voice.clone, voice.design, speech.tts | https://www.anchorterminal.com/tools/elevenlabs-voice-cloning.min.md |
| Inworld Instant Voice Cloning | C | 61.9 | voice.clone, voice.design, speech.tts | https://www.anchorterminal.com/tools/inworld-voice-cloning.min.md |
| Fish Audio Voice Cloning API | C | 61.5 | voice.clone, voice.design, speech.tts | https://www.anchorterminal.com/tools/fish-audio-voice-cloning.min.md |
| Ming-3.0-tts-0.5B | C | 59.3 | speech.tts, voice.clone, voice.design | https://www.anchorterminal.com/tools/ming-3-0-tts-0-5b.min.md |
| Inworld Realtime TTS-2 | C | 58.3 | speech.tts, voice.clone, voice.design | https://www.anchorterminal.com/tools/inworld-tts.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)