Photos, datasheets & media
Specs and sources stay open to everyone. Sign in to see the imagery and documents members contribute.
Stationary dual-arm robot for physical AI: 2x 7-DoF arms, 860 mm reach, 4 kg nominal (6.5 kg peak) payload per arm, 1 mm repeatability, 19.5 kg. MIT-licensed Python SDK with WebXR VR teleoperation and LeRobot bindings. From $7,999.
Base configuration · Listed Jul 2026
direct from Almond (Order Now). Options: Axol Mount +$199, Axol Cart +$999, Perception Kit (2x ZED X One S + ZED X Mini + ZED Box Orin NX) +$3,249; without grippers -$950.
~1 week, ships from San Francisco
Observed price, not an offer.
Save it, compare it or record that you have used it — with an account.
Axol is a dual-arm manipulation platform from Almond, a Y Combinator-backed hardware startup that designs and assembles the robot in San Francisco. It is aimed at teams collecting demonstration data and training manipulation policies for real deployments rather than lab demos. Each of the two arms has 7 degrees of freedom, 860 mm of reach and a wrist with full 180 degrees of pitch and yaw, which Almond says reduces shoulder singularities and enlarges the usable workspace. Peak payload is 6.5 kg per arm with a 4 kg nominal (sustained, non-thermally-throttled) rating. Repeatability is 1 mm and the robot itself weighs 19.5 kg. Motors run on a 500 Hz CAN bus exposed over a single locking USB-C connector, with XT90 power in from the included 1500 W (24 V, 62.5 A) supply. Construction is steel, aluminium and TPU with internally routed cabling, powder-coated matte black. Axol ships fully assembled with two 2-finger grippers whose tips are swappable (or the whole gripper can be replaced; gripper drawings and a STEP CAD model are published). It bolts down with four M8x12 bolts to a customer fixture, to the Axol Mount (a 400 mm 8080 T-slot column) or to the Axol Cart, a height-adjustable manually-rolled cart with leveling casters and a 60x60 cm footprint that fits through standard doors. The cart is manually repositioned - Axol has no driven mobile base. The optional Perception Kit adds 2x ZED X One S wrist cameras, 1x ZED X Mini head camera, all FAKRA GMSL2 cabling and mounts, and a ZED Box Orin NX 16GB for onboard compute. The robot body has GMSL2 ports and camera mounting holes for wrist and head cameras built in, so the kit is a bolt-on rather than a redesign. Software is an MIT-licensed open-source Python SDK and CLI (github.com/almond-bot/axol) covering low-level CAN motor control, a bimanual inverse-kinematics solver (pyroki/JAX), ZED camera streaming, LeRobot bindings for dataset collection and policy inference, a joint-tuning toolkit, a browser control panel, a diagnostics dashboard and a viser-based simulator. Teleoperation is VR-only: a WebXR app runs on any compatible headset (Meta Quest is documented, including a low-latency Quest-over-USB transport and remote teleop over Tailscale). There are no leader arms; gravity-compensation hand-guiding is offered as an alternative input. No ROS or ROS 2 integration is published.
8 other Bimanual Manipulation Platform products, closest hero specs and price first. Suggestions, not recommendations.
Almond Axol
This product
OpenArm 2
Enactic, Inc.
Galaxea R1 Lite
Galaxea Dynamics
Galaxea R1
Galaxea Dynamics
UFACTORY xArm Dual
UFACTORY
Dual UR5e
Universal Robots
This product is part of the Birdwave Atlas community.
Contribute evaluations, share integrations and help improve compatibility data.
Discussions, evaluations, integrations and updates
Discussions and reviews are for members
Sign in to read what members say about Almond Axol and to add your own evaluation, integration notes or spec fixes.
Sign in to join the discussionMIT-licensed Python SDK + CLI (github.com/almond-bot/axol): bimanual IK solver (pyroki/JAX), low-level CAN motor interface, WebXR VR teleop app, ZED camera streaming, LeRobot bindings, joint-tuning toolkit, web control panel, diagnostics dashboard, viser simulator. Public CAD: gripper.step, wrist_mount.step, head_mount.step plus mounting-hole and gripper PDF drawings. No URDF published.