# UltraX-0.6B-Preview

[Canonical model page](https://headlessthrones.com/models/openbmb-ultrax-0-6b-preview) · [Model library](https://headlessthrones.com/models)

Public catalog reference from Headless Thrones. Exact catalog ID: `openbmb-ultrax-0-6b-preview`. Catalog source-check date: 2026-10-23.

This brief contains the public catalog answers and any catalog guide, not private drafts, historical editorial revisions or independently measured rankings. The canonical HTML page may also display separately published editorial.

Current catalog answers

## What to know about UltraX-0.6B-Preview

AI-compiled from linked model documentation. This is a catalog reference, not an independent performance review. Catalog checked October 23, 2026.

[Exact model](https://headlessthrones.com/models/openbmb-ultrax-0-6b-preview#model-answer-identity) · [Access](https://headlessthrones.com/models/openbmb-ultrax-0-6b-preview#model-answer-access) · [Inputs & outputs](https://headlessthrones.com/models/openbmb-ultrax-0-6b-preview#model-answer-modalities) · [License](https://headlessthrones.com/models/openbmb-ultrax-0-6b-preview#model-answer-license) · [Model details](https://headlessthrones.com/models/openbmb-ultrax-0-6b-preview#model-answer-research) · [Weight details](https://headlessthrones.com/models/openbmb-ultrax-0-6b-preview#model-answer-repository)

### Which UltraX-0.6B-Preview does this page describe?

UltraX-0.6B-Preview is a model record from **OpenBMB** in the UltraX family. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview)

The recorded model identifier or checkpoint is `UltraX-0.6B-Preview`. Use that exact identifier when checking documentation or a provider listing. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview)

### How can you access UltraX-0.6B-Preview?

The catalog documents **Open weights** access  for UltraX-0.6B-Preview. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview)

As of the catalog check, this version was listed as a preview. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview)

Open weights means the model weights are distributed for use under their stated terms. It does not establish a hosted API, an open-source license, or a hardware requirement.

### What can UltraX-0.6B-Preview receive and generate?

Documented inputs: **text**. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview)

Documented outputs: **text**. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview)

Input support and output generation are separate capabilities. Reading an image does not establish image generation; a text output listing does not establish speech output. File sizes, formats and media limits depend on the documented implementation.

### What license is recorded for UltraX-0.6B-Preview?

The recorded license label is **apache-2.0**. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview)

Read the linked license and associated use terms for this exact version before redistribution, commercial deployment or fine-tuning. A label alone does not establish every permission, restriction or obligation, and it does not determine the terms of a separate hosted service.

### What distinguishes UltraX-0.6B-Preview?

These details come from the linked publisher material for this model. Reported capabilities and training descriptions are publisher statements, not independent test results.

**Task**: Predicts structured insertion, deletion and modification calls to refine web-crawled text. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/18d7c03559cea29ce4aac2741706b099819c5c68/README.md)

**Training method**: Full-parameter supervised fine-tuning for text-cleaning function calls. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/18d7c03559cea29ce4aac2741706b099819c5c68/README.md)

**Repository variants**: Three refinement variants are documented: instruction-guided/edit-weighted UltraX, a no-instruction ablation, and a preservation-weighted ablation with 60% keep\_all sampling. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/18d7c03559cea29ce4aac2741706b099819c5c68/README.md)

**Training sequence budget**: The training configuration uses a maximum sequence length of 20,480 tokens; this is not asserted as an inference context limit. [Source](https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/18d7c03559cea29ce4aac2741706b099819c5c68/README.md)

### What is documented about the UltraX-0.6B-Preview checkpoint?

This evidence refers to `openbmb/UltraX-0.6B-Preview` at recorded revision `18d7c03559cea29ce4aac2741706b099819c5c68`, checked October 23, 2026. [Repository metadata](https://huggingface.co/api/models/openbmb/UltraX-0.6B-Preview)

Repository task tag: `text-generation`. A task tag identifies a distribution category; it does not measure performance on that task. [Repository metadata](https://huggingface.co/api/models/openbmb/UltraX-0.6B-Preview)

Repository language labels: en. These labels are not independent evaluations of language quality. [Repository metadata](https://huggingface.co/api/models/openbmb/UltraX-0.6B-Preview)

Repository access gate: `false`. This records the repository's gate setting when checked; licensing and use terms still apply. [Repository metadata](https://huggingface.co/api/models/openbmb/UltraX-0.6B-Preview)

Read the retrieved model card for setup and qualifications. [Model card](https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/18d7c03559cea29ce4aac2741706b099819c5c68/README.md)

**Catalog qualifications:** Official publisher weight listing; availability refers to this distribution, not a guaranteed hosted API. Base, SFT and research variants are identified by their exact published names. [Source 1](https://huggingface.co/openbmb/UltraX-0.6B-Preview) · [Source 2](https://huggingface.co/api/models/openbmb/UltraX-0.6B-Preview) · [Source 3](https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/18d7c03559cea29ce4aac2741706b099819c5c68/README.md)

**Evidence still missing here:** context or input capacity, knowledge and training-data cutoffs, and tested hardware requirements. Missing documentation is not evidence that a feature is unavailable.

Headless Thrones has not independently benchmarked this model for this reference. Evaluate it on representative work before choosing it; these facts do not rank its accuracy, reliability, speed or suitability against another model.