Holo4: powering generalist computer-use agents

Holo4: powering generalist computer-use agents

Holo4:赋能通用计算机操作智能体

Holo4 is our new series of agentic models. It comes in two sizes: 27B dense and 35B-A3B Mixture of Experts. Both are available on the H Models API. We are also releasing an updated version of Holotron 3: Holotron4 Nano. Holo4 builds on our previous model and interacts with software through any available interface: GUIs, code, MCP and APIs. It scores well on academic benchmarks, but we built it for real business workflows. It was trained through supervised and reinforcement learning on a large set of environments and tasks, including those generated by our Agentic Task Factory.

Holo4 是我们全新的智能体模型系列。它包含两种规格:27B 稠密模型和 35B-A3B 混合专家模型(MoE)。两者均已在 H Models API 上线。同时,我们还发布了 Holotron 3 的更新版本:Holotron4 Nano。Holo4 基于我们之前的模型构建,能够通过任何可用接口与软件进行交互:包括图形用户界面(GUI)、代码、MCP 和 API。它在学术基准测试中表现优异,但我们开发它的初衷是为了应对真实的商业工作流。它通过监督学习和强化学习,在大量环境和任务中进行了训练,其中包括由我们的“智能体任务工厂”(Agentic Task Factory)生成的任务。

Models built for every interface

为所有接口而生的模型

Holo4 clicks and types on a screen, writes and runs its own code, and calls MCP or API tools. It uses whichever fits the task. Most agentic models are trained for one interface only: GUI-focused models are blind without a screen, while models that prefer tool calling are stuck in front of an application that has no API. Real work is not siloed that way, and a single business task can require combining these different approaches. Holo4 runs on desktops, on the web, on Android, in a code sandbox and against business APIs. It is the same model in each case and it is called the same way. You do not need to select a different model for each platform.

Holo4 可以在屏幕上进行点击和输入,编写并运行自己的代码,以及调用 MCP 或 API 工具。它会根据任务需求选择最合适的方式。大多数智能体模型仅针对单一接口进行训练:专注于 GUI 的模型在没有屏幕时会“失明”,而偏好工具调用的模型则在面对没有 API 的应用程序时束手无策。现实工作并非如此孤立,单一的商业任务往往需要结合多种方法。Holo4 可以在桌面端、网页端、Android 系统、代码沙箱以及商业 API 环境中运行。在每种情况下,它都是同一个模型,调用方式也完全一致。你无需为不同的平台选择不同的模型。

Holo4 models improve significantly over their Qwen base. Holo4 trails only the strongest closed models on long workflows: on OSWorld 2.0, Holo4 27B scores 61.7% against 81.8% for Opus 5.5, and Holo4 35B-A3B reaches 30.9%. However, it does so with orders of magnitude fewer parameters and at a much lower cost. We open-source every trajectory behind our scores on public benchmarks: replay each step at trajectories.hcompany.ai or download them from Hugging Face.

Holo4 模型相较于其 Qwen 基座模型有了显著提升。在长工作流任务中,Holo4 仅次于最顶尖的闭源模型:在 OSWorld 2.0 测试中,Holo4 27B 得分为 61.7%(Opus 5.5 为 81.8%),Holo4 35B-A3B 得分为 30.9%。然而,它是在参数量级大幅减少且成本显著降低的情况下实现这一表现的。我们开源了所有在公共基准测试中得分背后的轨迹数据:你可以在 trajectories.hcompany.ai 重放每一步操作,或从 Hugging Face 下载。

Competitive with the frontier, at a fraction of the cost

媲美前沿水平,成本仅为一小部分

On the hardest academic benchmarks for desktop control (OSWorld 2.0) and API use (AutomationBench), Holo4 competes with frontier models at a much lower cost per task.

在针对桌面控制(OSWorld 2.0)和 API 使用(AutomationBench)的最严苛学术基准测试中,Holo4 以远低于同类产品的单任务成本,与前沿模型展开竞争。

AI that does work

真正能干活的 AI

Trained on environments and tasks from our Agentic Task Factory, Holo4 models excel on professional software. The examples below show Holo4 27B alongside Qwen3.8 27B, its base model. Same prompt and harness for both models.

得益于在“智能体任务工厂”环境和任务中的训练,Holo4 模型在专业软件操作方面表现出色。以下示例展示了 Holo4 27B 与其基座模型 Qwen3.8 27B 的对比。两个模型使用相同的提示词和测试框架。

3D modeling · Eiffel tower 3D 建模 · 埃菲尔铁塔

Build a 3D model of the Eiffel Tower in FreeCAD, at a scale of 1 mm to 1 metre, centred on the origin and aligned to the X and Y axes. Work to this design. The tower is square in plan at every height, never round. Its half-width, measured from the central axis out to the corner, is 62.5 mm at ground level, 32.5 mm at height 57, 17.5 mm at height 115, and 9.35 mm at height 276. Between those heights the half-width follows a smooth curve that falls steeply near the ground and gently higher up, never a straight line. Four identical legs, one per quadrant, each a square column whose outer corner follows that profile. Each leg is 14 mm across at the ground and tapers to 4 mm at height 276. The legs stand apart from the ground up to the first platform, and converge as they rise. Nothing fills the space between them: the tower is open, and you can see straight through it from every side. Three platforms, each a solid square slab centred on the axis: 72 mm across and 4 mm thick at height 57; 40 mm across and 3 mm thick at height 115; 22 mm across and 3 mm thick at height 276. A mast from height 276 to 324, square, 8 mm across at its base tapering to 2 mm at the tip. Every part must be a closed solid with non-zero volume, and no part may fill the space between the legs.

在 FreeCAD 中构建一个埃菲尔铁塔的 3D 模型,比例为 1 毫米对应 1 米,以原点为中心,并与 X 和 Y 轴对齐。按照此设计进行操作:塔身在任何高度的平面图均为正方形,而非圆形。从中心轴到顶角的半宽测量值如下:地面处为 62.5 毫米,高度 57 处为 32.5 毫米,高度 115 处为 17.5 毫米,高度 276 处为 9.35 毫米。在这些高度之间,半宽遵循一条平滑曲线,靠近地面处下降陡峭,高处下降平缓,绝非直线。四条相同的腿,每个象限一条,每条腿为一个方形柱,其外角遵循上述轮廓。每条腿在地面处宽 14 毫米,在高度 276 处收窄至 4 毫米。腿部从地面到第一平台处是分开的,随着高度上升而汇聚。腿部之间的空间没有任何填充物:塔身是开放的,从任何一侧都可以直接看穿。三个平台,每个均为以轴线为中心的实心方形板:高度 57 处宽 72 毫米、厚 4 毫米;高度 115 处宽 40 毫米、厚 3 毫米;高度 276 处宽 22 毫米、厚 3 毫米。从高度 276 到 324 处设有一根桅杆,方形,底部宽 8 毫米,顶部收窄至 2 毫米。每个部件必须是具有非零体积的封闭实体,且任何部件不得填充腿部之间的空间。

3D modeling · H logo 3D 建模 · H 公司 Logo

Build a 3D model in FreeCAD of the H company logo: a solid filled disc beside a blocky sans-serif capital letter H, both extruded to the same thickness, the two shapes of similar height and set apart so they do not overlap, with the centre of the disc level with the middle of the H. Colour both shapes black.

在 FreeCAD 中构建 H 公司 Logo 的 3D 模型:一个实心填充圆盘位于一个块状无衬线大写字母 H 旁边,两者拉伸至相同厚度,两个形状高度相近且分开摆放互不重叠,圆盘中心与 H 的中部水平对齐。将两个形状均设为黑色。