# THX-01

> THX-01 is a 322M multilingual decision model from HAL-X AI, trained with reinforcement learning on a strictly proper scoring rule so its probabilities are calibrated. Send a state and typed questions (noul / choice / score) and get an answer with probabilities for each question in one forward pass of about 10 ms — not generated chat text. Post-trained in 18 languages, 100+ supported. Apache 2.0. Call POST /api/v1/decisions (POST /api/v1/systemone is the same handler).

**ID**: `hal-x/thx-01`  
**Creator**: hal-x  
**Category**: text  
**Context**: 1K tokens  
**Input**: $0  
**Output**: $0  
**Released**: 2026-10-11  
**Web page**: https://anyrouter.dev/model/hal-x/thx-01

**Input modalities**: text  
**Output modalities**: text  

**Capabilities**: systemone

**Supported parameters**: state, model, questions

## Usage

**Endpoint**: `POST https://anyrouter.dev/api/v1/decisions`

```bash
curl https://anyrouter.dev/api/v1/decisions \
  -H "Authorization: Bearer $ANYROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "state": "Help! My payouts have been failing for 3 days.",
  "model": "hal-x/thx-01",
  "questions": {
    "is_urgent": {
      "type": "noul",
      "instructions": "Does this convey urgency?"
    }
  }
}'
```

Decisions is not chat: send `state` and typed `questions` (noul / choice / score). Requires a BYOK key for TypeSafe, LLM Gateway, Vercel AI Gateway, AIHubMix, AutoJev, or Fastino.

## Providers

| Provider | Input | Output |
| --- | --- | --- |
| hal-x | $0 | $0 |

## Source

- Upstream docs: https://huggingface.co/doofz/THX-01