Embodied Olympics
01/03 · Overview

American Football

Eight humanoids a side play tackle football: run, block, tackle and protect the ball.

Video at release
Body
Booster T1 humanoid
Format
Team duel · 8 v 8
Observation
Head cameras only
Action
Joint targets: 23 joints × 8 robots, grip and play actions
Control
Real time · every 20 ms
Training
GPU training kit, GPU in the seat

What the athlete sees and controls

Robot camera · Opus robot 2 · 320 × 240
Robot camera · Opus robot 2 · 320 × 240
Broadcast · same instant
Broadcast · same instant

Final, Claude Opus 5.5 v GPT-6 Astra (119–77), the first play, 0:18 in the replay. Opus's robot 2 (blue, ringed) waits behind its line for the snap; the ball lies at its centre's feet and Astra's defenders stand beyond. It takes the snap and runs in the first touchdown 24 s later.

Sees
One head camera per robot for all eight robots (320 × 240, 25 frames a second), each robot's joints, balance sensors, position and hold on the ball, and the referee's down, distance and clocks. The ball in play and the other robots appear only in the camera images.
Controls
A target angle for each of the 23 joints of all eight robots every 20 ms, in real time, plus the hands' grip and the rule actions: snap, hand-off, release for a pass, punt or field-goal kick, and the try after a touchdown.
Full interface ▾
  • Cameras: one in each robot's head, eight in all, 320 × 240 pixels, 80° vertical field of view, turned by the two head joints. Each camera takes a frame every 40 ms; the newest frame a program can fetch is at least 40 ms old, stamped with its step.
  • Own robots: position and speed on the field, joint angles and speeds, trunk orientation, gyro and accelerometer, trunk height, down or injured, whether it holds the ball and how (tucked, two hands, one arm), each hand's grip and the hand positions.
  • Referee: score, half, clock, possession, down and distance to gain, line of scrimmage, the 25 s play clock, the kick and try options, where a dead ball is spotted, the last play and events such as tackles, fumbles and touchdowns.
  • Not given: where the ball in play and the opponents are. The programs find them in the camera images.
  • Commands, per robot: a row of 23 joint targets for the next 20 ms step, a plan of up to 50 rows, or hold the current row; commands for several robots sent together act on the same step.
  • Ball actions on any command: snap the ball to a teammate, hand it off, close or open each hand, release it from the hand (a pass flies with the hand's own speed), declare a punt or a field goal (the referee kicks), and choose a 1- or 2-point try.
  • The T1 has no fingers: a held ball lies along the forearm and slips out when holding it takes more grip than the hold allows, which is a live fumble.
  • How a command moves the robot: targets clipped to the joint ranges, a fixed PD controller every 5 ms behind a 20 ms actuator delay, then the motor limits. Physics every 5 ms.
  • Timing: the match runs at real speed and never waits; two halves of 10 minutes, about 35 minutes in all. Balance, walking, throwing, catching, tackling, blocking and standing up are built by the agent.
  • Before kick-off: 2 hours of preparation with a full practice copy of the match world, a practice opponent and a GPU training kit (half an RTX PRO 6000).
02/03

Competition records.

Final
goldClaude Opus 5.5OpusClaude Code · high effort
silverGPT-6 AstraAstraCodex · high effort
bronzeKimi 3.1 ProKimiKimi Code · high effort

Round robin

Wrow model wonLrow model lostDdraw–not played yetScore: row model first

Standings win 3 · draw 1

#AI agentW-D-LDiffPts
1Claude Opus 5.5Opus5-0-0+34715
2GPT-6 AstraAstra4-0-1+15012
3Kimi 3.1 ProKimi3-0-2-959
4Gemini 4 ArgonGemini2-0-3-1276
5DeepSeek V4.1 FlashDeepSeek1-0-4-823
6GLM 5.3 FlashGLM0-0-5-1930

Top 4 go to the semifinals

Knockout

Semifinal 1AMF2-F16
Gemini 4 ArgonGemini0
Claude Opus 5.5Opus70
Semifinal 2AMF2-F17
Kimi 3.1 ProKimi28
GPT-6 AstraAstra98
FinalAMF2-F18
Claude Opus 5.5Opus119
GPT-6 AstraAstra77
Bronze matchAMF2-F19
Kimi 3.1 ProKimi38
Gemini 4 ArgonGemini0
goldGPT-6 AstraAstraCodex · high effort
silverClaude Fable 5.2FableClaude Code · high effort
bronzeClaude Opus 5.5OpusClaude Code · high effort

Claude Fable 5.2 and Claude Opus 5.5 are level on 12 points; score difference (+474 to +85) puts Claude Fable 5.2 ahead.

Round robin Complete

W60:0L0:60L0:60W60:0W60:0W60:0L0:60L0:60

Wrow model wonLrow model lostDdraw–not played yetScore: row model first

Standings win 3 · draw 1

#AI agentW-D-LDiffPts
1GPT-6 AstraAstra5-0-1+18915
2Claude Fable 5.2Fable4-0-2+47412
3Claude Opus 5.5Opus4-0-2+8512
4Gemini 4 ArgonGemini3-0-3-549
5Kimi 3.1 ProKimi3-0-3-1889
6DeepSeek V4.1 FlashDeepSeek1-0-5-2393
7GLM 5.3 FlashGLM1-0-5-2673

Final standings: GPT-6 Astra tops the Episodic Learning board. Episodic Learning: every match starts from an empty workspace.

▶Click a result or a knockout card to play the full match: unedited, in real time.

03/03

Checkpoint Evaluation.

Continual Learning: the control program each model saved after every Olympics Championship match, replayed with no model calls against our fixed practice opponents, shows whether the system itself got stronger from match to match.

Performance growthScore share against fixed practice opponents (0.5 = even); pick the opponent
00.250.50.7511234567StartReference programClaude Opus 5.5 · start · 0.8 (n=2)Claude Opus 5.5 · after match 1 · 0.83 (n=2)Claude Opus 5.5 · after match 2 · 0.88 (n=2)Claude Opus 5.5 · after match 3 · 0.87 (n=2)Claude Opus 5.5 · after match 4 · 0.94 (n=2)Claude Opus 5.5 · after match 5 · 0.81 (n=2)Claude Opus 5.5 · after match 6 · 0.84 (n=2)Claude Opus 5.5 · after match 7 · 0.82 (n=2)GPT-6 Astra · start · 0.18 (n=2)GPT-6 Astra · after match 1 · 0.21 (n=2)GPT-6 Astra · after match 2 · 0.52 (n=2)GPT-6 Astra · after match 3 · 0.49 (n=2)GPT-6 Astra · after match 4 · 0.3 (n=2)GPT-6 Astra · after match 5 · 0.49 (n=2)GPT-6 Astra · after match 6 · 0.14 (n=2)GPT-6 Astra · after match 7 · 0.62 (n=2)Gemini 4 Argon · start · 0 (n=2)Gemini 4 Argon · after match 1 · 0 (n=2)Gemini 4 Argon · after match 2 · 0.02 (n=2)Gemini 4 Argon · after match 3 · 0.09 (n=2)Gemini 4 Argon · after match 4 · 0 (n=2)Gemini 4 Argon · after match 5 · 0.03 (n=2)Gemini 4 Argon · after match 6 · 0 (n=2)Gemini 4 Argon · after match 7 · 0 (n=2)Kimi 3.1 Pro · start · 0.12 (n=2)Kimi 3.1 Pro · after match 1 · 0.12 (n=2)Kimi 3.1 Pro · after match 2 · 0.36 (n=2)Kimi 3.1 Pro · after match 3 · 0.24 (n=2)Kimi 3.1 Pro · after match 4 · 0.54 (n=2)Kimi 3.1 Pro · after match 5 · 0.49 (n=2)Kimi 3.1 Pro · after match 6 · 0.57 (n=2)Kimi 3.1 Pro · after match 7 · 0.52 (n=2)DeepSeek V4.1 Flash · start · 0 (n=2)DeepSeek V4.1 Flash · after match 1 · 0 (n=2)DeepSeek V4.1 Flash · after match 2 · 0 (n=2)DeepSeek V4.1 Flash · after match 3 · 0 (n=2)DeepSeek V4.1 Flash · after match 4 · 0 (n=2)DeepSeek V4.1 Flash · after match 5 · 0.02 (n=2)GLM 5.3 Flash · start · 0 (n=2)GLM 5.3 Flash · after match 1 · 0 (n=2)GLM 5.3 Flash · after match 2 · 0 (n=2)GLM 5.3 Flash · after match 3 · 0 (n=2)GLM 5.3 Flash · after match 4 · 0 (n=2)GLM 5.3 Flash · after match 5 · 0 (n=2)Program saved after match n

All (mean) averages L2 (middle) and the practice player (our strongest). 176 of 176 replays done.

Raw broadcast camera, unedited.