HomeLoop
Agents coordinate robots and connected devices to investigate household alerts. A program checks their actions and whether they resolve the problem.
- Made by
- Ramy ElMallah and colleagues
- Released
- November 2026
- Kind
- Simulated robots
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Agent task
- Measures
- Capability
- Language
- English
- In and out
- Text, Images, Time series → Actions
- Interaction
- Agentic
- Environment
- Isaac Sim, APIs and tools
- Refresh
- One fixed set
- Run by
- A host
- Strongest model tested
- GPT-6 Astra #4 on the capability index
- Last model added
- September 2026
- Scores refreshed
- November 20th, 2026
GPT-6 Astra leads with 61.9% among models released by September 2026
Best score per model from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in September 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6.1 Sol
- and 15 more
- 1 GPT-6 Astra medium reasoning 61.9%
- 2
Gemini Robotics ER 2 medium reasoning 51.4%
- 3 Claude Sonnet 5 adaptive reasoning 51.2%
- 4
Gemma 4 31B IT default reasoning 42.9%
- 5 GPT-6 Luna medium reasoning 39%
- 6
Qwen3.5-9B default reasoning 24%
Source: The HomeLoop paper
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6 Astra | 61.9% | medium reasoning | Sep 2026 | Official leaderboard |
| 2 | | 51.4% | medium reasoning | Jul 2026 | Official leaderboard |
| 3 | Claude Sonnet 5 | 51.2% | adaptive reasoning | Jun 2026 | Official leaderboard |
| 4 | | 42.9% | default reasoning | Apr 2026 | Official leaderboard |
| 5 | GPT-6 Luna | 39% | medium reasoning | Sep 2026 | Official leaderboard |
| 6 | | 24% | default reasoning | Feb 2026 | Official leaderboard |
Rankings by part
We read 6 parts from the benchmark’s paper, each ranking models on one part of the test, such as Hazards, Requests and Entry. GPT-6 Astra leads 3 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Hazards
6 models we track have a score here.
- 1
Gemma 4 31B IT ↑3 64%
- 2 Claude Sonnet 5 ↑1 64%
- 3
Gemini Robotics ER 2 ↓1 57%
- 4 GPT-6 Astra ↓3 43%
- 5
Qwen3.5-9B ↑1 29%
- 6 GPT-6 Luna ↓1 29%
Requests
6 models we track have a score here.
- 1 GPT-6 Astra 83%
- 2 GPT-6 Luna ↑3 75%
- 3
Gemma 4 31B IT ↑1 67%
- 4
Gemini Robotics ER 2 ↓2 58%
- 5 Claude Sonnet 5 ↓2 50%
- 6
Qwen3.5-9B 33%
Entry
6 models we track have a score here.
- 1 Claude Sonnet 5 ↑2 67%
- 2
Gemini Robotics ER 2 67%
- 3 GPT-6 Luna ↑2 67%
- 4
Qwen3.5-9B ↑2 33%
- 5
Gemma 4 31B IT ↓1 33%
- 6 GPT-6 Astra ↓5 33%
Wellbeing
6 models we track have a score here.
- 1 GPT-6 Astra 100%
- 2 Claude Sonnet 5 ↑1 75%
- 3
Gemini Robotics ER 2 ↓1 75%
- 4
Gemma 4 31B IT 50%
- 5
Qwen3.5-9B ↑1 25%
- 6 GPT-6 Luna ↓1 25%
Chores
6 models we track have a score here.
- 1 GPT-6 Astra 50%
- 2
Qwen3.5-9B ↑4 0%
- 3
Gemma 4 31B IT ↑1 0%
- 4 Claude Sonnet 5 ↓1 0%
- 5
Gemini Robotics ER 2 ↓3 0%
- 6 GPT-6 Luna ↓1 0%
Alert groups
6 models we track have a score here.
- 1
Gemini Robotics ER 2 ↑1 38%
- 2 GPT-6 Astra ↓1 25%
- 3 Claude Sonnet 5 13%
- 4
Qwen3.5-9B ↑2 0%
- 5
Gemma 4 31B IT ↓1 0%
- 6 GPT-6 Luna ↓1 0%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The HomeLoop paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official HomeLoop leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Robotics
- ERQA In use 400 multiple-choice visual questions from robot and egocentric footage on spatial, trajectory, action and state reasoning. Accuracy.
Gemini Robotics ER 2 78.5%
- SimplerEnv In use Real-to-sim replicas of Google Robot and WidowX/Bridge setups. Policies scored by success rate on pick, move, drawer and Bridge tasks. Praxis-1 80%
- RoboLab In use Zero-shot DROID Franka policies in Isaac Lab on 120 tabletop tasks at three instruction levels, 10 episodes each. Success rate. Praxis-1 46.2%
- RoboDojo In use Generalist manipulation policies run 42 Isaac Sim tasks and 18 remote real-robot tasks. Scored by task score and success rate. GPT-6 Astra 27.6%
- Progress classification (Google DeepMind) In use Models assign each frame of a robot video to one of five task-progress bands from 0 to 100 percent. Scored by accuracy.
Gemini Robotics ER 2 57.4%
- Human proximity (Google DeepMind robotics) In use Models judge from robot camera views whether a human is within 1 meter, a spatial awareness check for safe robot behavior. Scored by accuracy.
Gemini Robotics ER 2 93%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (6 entries).
- Pass Models scored 6 models we track.
- Pass Newest score The newest is from September 2026, for GPT-6 Luna.
- Note Frontier models Only 1 of the top 10 has a score. The strongest tested is GPT-6 Astra (#4).
- Pass Headroom The best is 61.9%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 6 rows for models we track at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.