Unitree H1 Dedicated Computing Module
Genuine Unitree Jetson Orin NX external compute for the H1 — one board assembly delivering 100 TOPS of CUDA compute, back-mounted in its dedicated protective bracket, up to two modules per robot.
- Unitree H1
About the Unitree H1 Dedicated Computing Module
The Unitree H1 dedicated computing module is the external AI compute upgrade for Unitree's first full-size humanoid: one or two NVIDIA Jetson Orin NX computing board assemblies in a protective enclosure that mounts on the robot's back and turns a CPU-only machine into a CUDA-equipped platform. It comes in two versions — a 100 TOPS module built around a single Orin NX assembly, and a 200 TOPS module that pairs two of them in a single enclosure for heavier perception stacks.
A standard H1 ships with two x86 computers — an Intel Core i5 running the platform and a Core i7 reserved for your code — but neither is a GPU. When a stack outgrows CPU compute — multi-camera perception, learned navigation, VLA-class models, real-time sensor fusion — the Unitree H1 dedicated computing module adds the missing layer without opening the body: it hangs on the back in its dedicated high-strength protective bracket, the internal computers keep their roles, and it moves to another H1 as a hardware swap.
Choose the 100 TOPS version for demanding-but-standard autonomy and machine-vision workloads — an H1 can carry up to two of them — or the 200 TOPS dual version when several large models must run at once. The Unitree H1 dedicated computing module arrives fully assembled with the bracket included — one unit, one mount, no integration project.
What the Unitree H1 Dedicated Computing Module Adds to a Robot That Already Has an i7
Unitree's H1 specification lists the standard compute as an Intel Core i5 for the platform and a Core i7 for user development, with Jetson-class GPU options named as extras. That pair handles locomotion, control and housekeeping well, but it is CPU compute — there is no CUDA GPU in the standard machine. The Unitree H1 dedicated computing module closes exactly that gap: NVIDIA Jetson Orin NX assemblies mounted on the robot's back, adding 100 or 200 TOPS of GPU-accelerated AI performance for camera-heavy perception, deep-learning inference and multi-sensor fusion.
Because the module lives outside the body, nothing inside is disturbed: the internal computers keep their roles, and the GPU layer rides on top. When an H1 is reassigned, or the module needs to move to another unit in the fleet, an external mount makes that a hardware swap rather than a teardown.
100 TOPS or 200 TOPS: Which Version Should You Choose?
Both versions of the Unitree H1 dedicated computing module mount the same way and fit the same robot — the difference is how much CUDA compute rides on the back, and in how many housings. The 100 TOPS module carries a single Jetson Orin NX assembly in its own protective bracket, and an H1 can carry up to two of them. The 200 TOPS module pairs two Orin NX assemblies in a single enclosure, so one mount delivers the same two-board compute — and one module is all an H1 takes.
- Choose the 100 TOPS module for demanding-but-standard autonomy: multi-camera perception, SLAM, learned navigation and edge inference fit comfortably on a single Orin NX — and a second module can be added later, up to two per robot.
- Choose the 200 TOPS dual module when the robot must run heavyweight pipelines concurrently — VLA-class models, dense multi-sensor fusion, or AI running alongside a second heavy application — all in a single enclosure.
- Size for the model you plan to run in twelve months, not the one you run today — an undersized module is the wrong saving.
The Dedicated Computing Module Is Not the AGX Module
The H1 offers two external GPU paths, and they are not interchangeable. The Unitree H1 dedicated computing module is built on NVIDIA Jetson Orin NX — 100 TOPS per board, up to two boards on one H1. The AGX module family is built on the larger Jetson AGX Orin — 275 or 550 TOPS — and exists as its own product family with its own part numbers. The two are complements, not competitors: a fully loaded H1 pairs two Orin NX boards with a 550 TOPS AGX module for roughly 750 TOPS of combined AI compute. Match the module to the workload — Orin NX for standard perception stacks, AGX for the heaviest multi-model rigs.
Built for the H1's Back
The Unitree H1 dedicated computing module is an H1 part: it mounts on the H1's back via its dedicated high-strength protective bracket, and the per-robot limits above are H1 limits. The newer H1-2 is a different body — it takes its own internal computing boards rather than this external module, so this module is not the part for an H1-2.
Unitree's other platforms carry their own compute parts: the B2 family has its Jetson Orin NX computing board and AGX module, and the G1 takes AGX Orin and AGX Thor backplane docks. This page's module is the H1 version.
Specifications
| 100 TOPS | 200 TOPS | |
|---|---|---|
| Module | 1 × NVIDIA Jetson Orin NX computing board assembly | 2 × NVIDIA Jetson Orin NX computing board assemblies (dual, single enclosure) |
| AI performance | 100 TOPS | 200 TOPS (2 × 100 TOPS) |
| CPU | 8-core Arm Cortex-A78AE (NVIDIA module spec) | 2 × 8-core Arm Cortex-A78AE (NVIDIA module spec) |
| GPU | NVIDIA Ampere architecture with Tensor Cores (NVIDIA module spec) | 2 × NVIDIA Ampere architecture with Tensor Cores (NVIDIA module spec) |
| Memory | 16GB (NVIDIA module spec) | 2 × 16GB — 32GB total (NVIDIA module spec) |
| Mounting | External — back-mounted, dedicated protective bracket included | External — back-mounted, dedicated protective bracket included |
| Per robot | Up to 2 modules | 1 module |
| Fits | Unitree H1 | Unitree H1 |
| In the box | 1 × Unitree H1 Dedicated Computing Module — 100 TOPS (single Jetson Orin NX board assembly, dedicated protective bracket included) | 1 × Unitree H1 Dedicated Computing Module — 200 TOPS (dual: 2 × Jetson Orin NX computing board assemblies, dedicated protective bracket included) |
Shipping, Quality & Module Care
The module contains no lithium cells, so it ships as standard express freight rather than battery-class cargo — worldwide via DHL or FedEx, with the transit time confirmed in your quote.
Every Unitree H1 dedicated computing module is 100% quality-inspected before it ships and packed as protected electronic cargo, so it arrives in perfect working order. From there it is sensitive electronics like any other: dry hands, gentle with the connectors, and keep the module dry once it is on the robot.
Last updated: October 2026.
Frequently Asked Questions
Can't find what you're looking for? Contact our
