> ## Documentation Index
> Fetch the complete documentation index at: https://docs.abliteration.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# abliteration.ai models

> abliteration.ai serves three unrestricted reasoning models: abliterated-model (multimodal, 256K context), abliterated-model-large-v2 (text-only, 1M context, GLM-5.3), and abliterated-model-large (text-only, 1M context, GLM-5.2).

All three models are uncensored, think before answering by default (see [thinking & reasoning effort](/capabilities/thinking)), stream, and call tools.

* **`abliterated-model`** — the general-purpose default. Multimodal with image inputs, 256K context.
* **`abliterated-model-large-v2`** — the default large model. Derived from the open-weight GLM-5.3 model, further abliterated and fine-tuned by Abliteration AI. Text-only, 1M context, for harder reasoning and evaluation workloads. Runs in three reasoning modes — **low**, **high**, and **max**.
* **`abliterated-model-large`** — the previous large model, derived from the open-weight GLM-5.2 model, further abliterated and fine-tuned by Abliteration AI. Text-only, 1M context.

Request a model by passing its identifier in the `model` field:

```json theme={"system"}
{ "model": "abliterated-model", "messages": [...] }
```

## Limits

| Model | Context | Max output |
| - | - | - |
| `abliterated-model` | 262,144 | 262,134 |
| `abliterated-model-large-v2` | 1,000,000 | 999,990 |
| `abliterated-model-large` | 1,000,000 | 999,990 |

Context is the combined input + output budget; max output is the largest completion the model will return, reachable only when the prompt is small.

## Capabilities

| Capability | Supported |
| - | - |
| Chat completions / Responses | ✓ |
| Streaming | ✓ |
| Tool / function calling | ✓ |
| Structured outputs (JSON mode, JSON Schema) | ✓ |
| Reasoning effort control | ✓ |
| Disable reasoning | ✓, except Large V2 maps disable requests to hidden low reasoning |
| Hide reasoning | ✓ |
| Reasoning trace in responses | ✓ |
| Images (`abliterated-model` only) | ✓ |
| Video inputs | ✗ |
| Web search | ✓ |
| Web fetch (Anthropic only) | ✓ |

See [compatibility matrix](/compatibility-matrix) for the full per-endpoint breakdown.

<Note>
  Image inputs apply to `abliterated-model` only. `abliterated-model-large` and `abliterated-model-large-v2` are text-only and reject image content with a `400` error. Video inputs are not supported by any model.
</Note>

## Model listing

The live list is returned by `GET /v1/models`:

```sh theme={"system"}
curl https://api.abliteration.ai/v1/models \
  -H "Authorization: Bearer $ABLIT_KEY"
```

## Pricing

Usage-based, billed on total tokens (input + output). See [pricing](/pricing) for per-token rates by model, or the [pricing page](https://abliteration.ai/pricing) for plans and credit packs.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.