# Liquid d1.2

> Liquid d1.2 is Liquid AI's faster hosted decision model. It returns calibrated probabilities instead of generated text: send a state and typed questions in one call, with zero generated tokens. Use it for classification, routing, scoring and extraction.

**ID**: `liquid/d1.2`  
**Creator**: liquid  
**Category**: text  
**Context**: 66K tokens  
**Pricing**: $0.02 / $0 per 1M · billed by your provider (BYOK)  
**Released**: 2026-11-16  
**Web page**: https://anyrouter.dev/model/liquid/d1.2

**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": "liquid/d1.2",
  "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 |
| --- | --- | --- |
| liquid-byok | $0.02/M (list) | $0 (list) |

BYOK routes are billed by your own provider key — AnyRouter charges $0. Rates marked (list) are the provider's published list price.

## Model card

- Native context: 65536 tokens [1]

### Strengths [1,3]

- Faster hosted decision model, priced at less than half of d1.2-ultra.
- Typed answers read values, find passages, locate objects and estimate ranges.

### Known limits [1]

- Returns calibrated answers, not generated text; it is not a chat model.
- 65,536-token input limit per question and a 250,000-byte text limit.

### Use cases [2,1]

- Classification and routing
- Scoring and moderation gates
- Extracting typed values from documents

### API notes [1,2,3,4]

- Hosted on Liquid's Decisions API at POST /decisions/v1/systemone with {model, state, questions}.
- Billed on input tokens only at $0.02 per million; output tokens are free.
- Adds text, number, date, spans, box/boxes and ranged number answers on top of Noul, Choice and Score.

### Links

- announcement: https://x.com/liquidai/status/2122346077231923698
- docs: https://docs.liquid.ai/lfm/models/d1
- pricing: https://www.liquid.ai/pricing

### Sources

1. [d1.2 — Liquid Docs](https://docs.liquid.ai/lfm/models/d1) (accessed 2026-11-17)
2. [Decision Models — Liquid Docs](https://docs.liquid.ai/lfm/models/decision-models) (accessed 2026-11-17)
3. [Liquid AI launches d1.2 and d1.2-ultra](https://x.com/liquidai/status/2122346077231923698) (accessed 2026-11-17)
4. [Liquid AI pricing](https://www.liquid.ai/pricing) (accessed 2026-11-17)

## Source

- Upstream docs: https://docs.liquid.ai/lfm/models/d1
