Seeing What Execution Needs
Agents choose camera views that help a robot complete a tabletop task. The evaluator runs the action and checks whether the requested state holds.
- Released
- November 2026
- Kind
- Simulated robots
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Agent task
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- A simulator
- Refresh
- One fixed set
- Run by
- A host
- Strongest model tested
- Grok 4.6 #29 on the capability index
- Last model added
- August 2026
- Scores refreshed
- November 20th, 2026
VBM+AIBR tops the leaderboard with 98.9% Grok 4.3 leads with 42.9% among models released by August 2026
Best score per entry 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. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Grok 4.3, at 42.9%. This view shows only models we track, each at its best.
The newest model with a score here came out in August 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 Astra
- and 16 more
- 1 VBM+AIBR C.2: Same-object functional OOD 98.9%
- 2 View Belief Map C.2: Same-object functional OOD 98.8%
- 3 Grok 4.3 Strict-clean foundation-model view scoring 42.9%
- 4 Grok 4.6 Strict-clean foundation-model view scoring 42.5%
- 5
Amazon Nova Pro Strict-clean foundation-model view scoring 40.3%
- 6 Qwen3-VL-8B Strict-clean foundation-model view scoring 39.4%
- 7
Gemma 3 27B Strict-clean foundation-model view scoring 32.5%
- 8
Llama 4 Maverick Strict-clean foundation-model view scoring 28.3%
- 9
Sonar Pro Strict-clean foundation-model view scoring 26%
Source: The Seeing What Execution Needs paper
- 1 Grok 4.3 Strict-clean foundation-model view scoring 42.9%
- 2 Grok 4.6 Strict-clean foundation-model view scoring 42.5%
- 3
Amazon Nova Pro Strict-clean foundation-model view scoring 40.3%
- 4
Gemma 3 27B Strict-clean foundation-model view scoring 32.5%
- 5
Llama 4 Maverick Strict-clean foundation-model view scoring 28.3%
- 6
Sonar Pro Strict-clean foundation-model view scoring 26%
Source: The Seeing What Execution Needs paper
| # | Entry | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | VBM+AIBR | 98.9% | C.2: Same-object functional OOD | Official leaderboard | |
| 2 | View Belief Map | 98.8% | C.2: Same-object functional OOD | Official leaderboard | |
| 3 | Grok 4.3 | 42.9% ±2.5 | Strict-clean foundation-model view scoring | Apr 2026 | Official leaderboard |
| 4 | Grok 4.6 | 42.5% ±2.5 | Strict-clean foundation-model view scoring | Aug 2026 | Official leaderboard |
| 5 | | 40.3% ±2.5 | Strict-clean foundation-model view scoring | Dec 2024 | Official leaderboard |
| 6 | Qwen3-VL-8B | 39.4% ±2.5 | Strict-clean foundation-model view scoring | Official leaderboard | |
| 7 | | 32.5% ±2.4 | Strict-clean foundation-model view scoring | Mar 2025 | Official leaderboard |
| 8 | | 28.3% ±2.3 | Strict-clean foundation-model view scoring | Apr 2025 | Official leaderboard |
| 9 | | 26% ±2.2 | Strict-clean foundation-model view scoring | Jan 2025 | Official leaderboard |
| # | Model | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Grok 4.3 | 42.9% ±2.5 | Strict-clean foundation-model view scoring | Apr 2026 | Official leaderboard |
| 2 | Grok 4.6 | 42.5% ±2.5 | Strict-clean foundation-model view scoring | Aug 2026 | Official leaderboard |
| 3 | | 40.3% ±2.5 | Strict-clean foundation-model view scoring | Dec 2024 | Official leaderboard |
| 4 | | 32.5% ±2.4 | Strict-clean foundation-model view scoring | Mar 2025 | Official leaderboard |
| 5 | | 28.3% ±2.3 | Strict-clean foundation-model view scoring | Apr 2025 | Official leaderboard |
| 6 | | 26% ±2.2 | Strict-clean foundation-model view scoring | Jan 2025 | Official leaderboard |
Rankings by part
We read 5 parts from the benchmark’s paper, each ranking models on one part of the test, such as Tamp, Insert and Place. Grok 4.3 leads 2 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.
Tamp
6 models we track have a score here.
- 1
Sonar Pro ↑5 98%
- 2
Amazon Nova Pro ↑1 97.7%
- 3 Grok 4.6 ↓1 97.7%
- 4
Llama 4 Maverick ↑1 97.3%
- 5 Grok 4.3 ↓4 67%
- 6
Gemma 3 27B ↓2 63.3%
Insert
6 models we track have a score here.
- 1 Grok 4.3 10.7%
- 2
Amazon Nova Pro ↑1 0%
- 3
Sonar Pro ↑3 0%
- 4
Gemma 3 27B 0%
- 5
Llama 4 Maverick 0%
- 6 Grok 4.6 ↓4 0%
Place
6 models we track have a score here.
- 1
Amazon Nova Pro ↑2 99.7%
- 2
Gemma 3 27B ↑2 99.3%
- 3 Grok 4.3 ↓2 99.3%
- 4 Grok 4.6 ↓2 97.3%
- 5
Llama 4 Maverick 43.7%
- 6
Sonar Pro 14.3%
Serve
6 models we track have a score here.
- 1 Grok 4.3 37.3%
- 2
Sonar Pro ↑4 16.7%
- 3
Amazon Nova Pro 4.30%
- 4 Grok 4.6 ↓2 3.70%
- 5
Llama 4 Maverick 0.30%
- 6
Gemma 3 27B ↓2 0%
Grind
6 models we track have a score here.
- 1 Grok 4.6 ↑1 13.7%
- 2
Sonar Pro ↑4 1.00%
- 3
Amazon Nova Pro 0%
- 4
Gemma 3 27B 0%
- 5
Llama 4 Maverick 0%
- 6 Grok 4.3 ↓5 0%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The Seeing What Execution Needs 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 Seeing What Execution Needs 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 (9 entries).
- Pass Models scored 6 models we track, plus 3 other entries on its leaderboard.
- Pass Newest score The newest is from August 2026, for Grok 4.6.
- Note Frontier models None of the top 10 has a score. The strongest tested is Grok 4.6 (#29).
- Pass Headroom The best is 42.9%, 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 and 3 other entries 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.