# three.ws Avatar Optimize Batch

> three.ws Avatar Optimize Batch is a paid API for AI agents from three.ws, paid per call via x402, $0.001/call, status unknown (last checked 2026-10-11).

Analyzes up to 50 3D avatars in batch and returns optimization suggestions including compression recommendations, file size metrics, and severity-ranked warnings.

## Facts

- Endpoint: POST https://three.ws/api/x402/avatar-optimize-batch?utm_source=zero.xyz
- Price: $0.001/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-11
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/three-ws-avatar-optimize-batch-357f5e0c
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_0u2zfAJo6cmsOCzCotVQv

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability three-ws-avatar-optimize-batch-357f5e0c -d '<json body>'
```

Example prompt: Can you run a batch optimization audit on my top 30 three.ws avatars and tell me which ones have critical issues and what the best fix suggestions are?

## When to prefer this

Use this endpoint when you need to audit a batch of 3D avatars for performance and optimization issues on the three.ws platform. It is the right choice when you want prioritized, actionable recommendations across many avatars at once rather than analyzing a single avatar. Prefer this over manual inspection when managing a large library of 3D AI agents and need to triage critical vs minor issues efficiently.

## Known failure modes

- limit out of range (must be 1–50) returns validation error
- no avatars found in account returns empty list
- authentication/payment failure via x402 returns 402 Payment Required
- server-side analysis timeout for very large avatar assets

## How this service works

Make a 3D AI agent from a selfie in 60 seconds. Drop one HTML tag to embed it on any website. Agents think on IBM watsonx.ai (Granite); optionally charge per chat with built-in micropayments.

## Output

Returns a list of analyzed avatars each with their UUID, name, file size in bytes, info/warn/critical issue counts, and a top optimization suggestion (e.g. 'Apply Draco compression to reduce geometry size by ~60%'). Also returns aggregate totals: number analyzed, total file size, and rolled-up info/warn/critical counts across all avatars.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "limit": {
   "type": "integer",
   "default": 50,
   "maximum": 50,
   "minimum": 1,
   "description": "Number of top avatars to analyze (1–50)."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "avatars": [
   {
    "id": "uuid",
    "name": "My Avatar",
    "info_count": 5,
    "size_bytes": 2100000,
    "warn_count": 3,
    "critical_count": 1,
    "top_suggestion": "Apply Draco compression to reduce geometry size by ~60%"
   }
  ],
  "analyzed": 50,
  "info_count": 91,
  "warn_count": 38,
  "critical_count": 12,
  "total_size_bytes": 82000000
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/three-ws-avatar-optimize-batch-357f5e0c/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from three.ws](https://www.zero.xyz/host/three.ws/llms.txt)