# Smallest.ai Instant Voice Clone (slim)

> Smallest.ai's instant voice cloning API creates a reusable voice for its Lightning text-to-speech models from a 5 to 15 second sample in one multipart POST, returning the voice ID and sample clips of the cloned voice.

- Full: https://www.anchorterminal.com/tools/smallest-voice-clone.md (~7,900 tokens) · this version ~1,830 tokens · JSON https://www.anchorterminal.com/tools/smallest-voice-clone.json · canonical https://www.anchorterminal.com/tools/smallest-voice-clone
- 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/smallest-voice-clone#pricing · reviews https://www.anchorterminal.com/tools/smallest-voice-clone#reviews
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

**C · 58/100 · rank #768 of 1486 · #8 in Voice cloning & custom voices · not agent-ready · confidence medium**


Assessment: One multipart POST clones a voice from 5 to 15 seconds of audio and returns sample clips, with dated deprecation notices and current changelog entries. The endpoint still defaults to a model whose clones stop working on 30 November 2026, there's no consent field or clone price, and the status feed shows frequent short API outages.

## Facts

- Kind: Model API · vendor: Smallest, Inc. · category: Voice cloning & custom voices · legal entity: Smallest, Inc. · provenance 87/100
- Endpoint: `https://api.smallest.ai/waves/v1/voice-cloning` (HTTP)
- Auth: API key · pricing: Pay per use · x402: no · licence: Proprietary service under Smallest's Terms of Service. The `smallestai` 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 18:08 UTC · uptime 30 days 100.0% (198 probes) · p50 361 ms · p95 430 ms over 24 hours
- Vendor status page: not read, no machine-readable status found (2026-11-07 04:48 UTC)
- Surface listed: The instant cloning endpoint `POST /waves/v1/voice-cloning` on api.smallest.ai, read from docs.smallest.ai (guide, API reference, limits, errors, deprecations, changelog), the pricing, terms, privacy and prohibited-use pages, status.smallest.ai and PyPI on 6 November 2026. No account was created and no call was made
- Sample length: 5 to 15 seconds, under 5 MB, as MP3, WAV, MP4 or WebM, one file per clone
- Endpoints: `POST /waves/v1/voice-cloning` (multipart, synchronous, returns sample clips), `GET /waves/v1/voice-cloning` (lists every clone with `modelIds`), `DELETE /waves/v1/lightning-large` (deletes a clone). The older `POST /waves/v1/lightning-large/add_voice` is deprecated
- Models: Clones are made for one pool, `lightning-v3.1` (default, retiring on 30 November 2026) or `lightning-v3.1-pro`, and must be used with the matching TTS model. `lightning-v2` is rejected
- Consent and verification: No consent or verification field in the API. The Terms let Smallest ask for documentation that the speaker authorised the clone and remove voices without it. The prohibited use policy bars unauthorised impersonation
- Voice ownership: The Terms say the customer owns its Input and Smallest assigns its rights in the Output, but the customer may not sell or share Output with third parties for commercial benefit
- Limits: One concurrent TTS request per account on the self-serve plan, 100 TTS requests a minute and 450 million characters a month per the pricing table. A per-organisation clone limit returns `clone limit exceeded, consider upgrading`, with no number published
- Regions: India (Mumbai, api.india.smallest.ai) and USA (Oregon, api.us.smallest.ai), routed by proximity, with region-pinned hostnames
- Data retention: Zero Data Retention and retention windows of 24 hours to 90 days are Enterprise options. No retention period for cloning samples is published
- Legal entity: Smallest, Inc., 311 California Street, Suite 320, San Francisco, under California law per the Terms
- Prices: Speech from a clone, Lightning v3.1 $17.50 per 1M characters; Speech from a clone, Lightning v3.1 Pro $19.50 per 1M characters

- Scores: Reliability 65, Performance pending, Schema & documentation 70, Agent ergonomics 62, Security & auth 48, Payments & pricing 20, Task success pending, Maintenance & community 72, Transparency & trust 72 · total over the 7 assessed categories
- Why: Reliability, Graded on the hosted cloning API with the hosted lines. · Schema & documentation, The API reference pages are generated with typed fields, enums and error schemas, but no OpenAPI file was found linked from the docs or llms… · Agent ergonomics, The clone response is a compact object with the voice ID, status and signed URLs to sample clips rather than inline audio (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 naming Smallest, Inc.
- Sources: 16, open questions: 7, both in the full twin
- Capabilities: voice.clone, speech.tts
- JSON: https://www.anchorterminal.com/api/v1/tools/smallest-voice-clone.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/smallest-voice-clone.svg` or a link to https://www.anchorterminal.com/tools/smallest-voice-clone from a page on smallest.ai or one of its subdomains, 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/9WtSXv26WE; support@smallest.ai; https://docs.smallest.ai/; more in the full twin

## Before you call it

1. Always pass `model=lightning-v3.1-pro`. The default `lightning-v3.1` clones stop working on 30 November 2026
2. Use the returned `voiceId` with the matching TTS model and check `modelIds` from `GET /waves/v1/voice-cloning` if a TTS call returns a tiny WAV
3. Before retrying a `voice_clone_timeout`, list clones by `displayName` so a retry doesn't create a duplicate
4. Send the server API key, not an access token. The cloning endpoint refuses short-lived tokens
5. Keep TTS calls from a clone to one at a time on the self-serve plan and back off on 429

## Connect

```bash
pip install smallestai
```

```bash
curl -X POST "https://api.smallest.ai/waves/v1/voice-cloning" \
  -H "Authorization: Bearer $SMALLEST_API_KEY" \
  -F "displayName=my-custom-voice" \
  -F "file=@/path/to/sample.wav" \
  -F "model=lightning-v3.1-pro" \
  -F "language=en"
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| Speechify API Voice Cloning | BB | 76.9 | voice.clone, speech.tts | https://www.anchorterminal.com/tools/speechify-voice-cloning.min.md |
| ElevenLabs Voice Cloning and Voice Design API | BB | 76.8 | voice.clone, speech.tts | https://www.anchorterminal.com/tools/elevenlabs-voice-cloning.min.md |
| HeyGen Voice | BB | 73.6 | speech.tts, voice.clone | https://www.anchorterminal.com/tools/heygen-voice.min.md |
| Soniox Voice Cloning | B | 68.7 | voice.clone, speech.tts | https://www.anchorterminal.com/tools/soniox-voice-cloning.min.md |
| Inworld Instant Voice Cloning | C | 61.9 | voice.clone, speech.tts | https://www.anchorterminal.com/tools/inworld-voice-cloning.min.md |


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