Robotics and CAD / Benchmark leaderboard
CALVIN
Long-horizon chains of language-conditioned robot manipulation.
No published company result yet.
This benchmark is in the research directory. Its task package, adapter and grading protocol need qualification before a hosted run can be offered.
Arrange an evaluation for your company →What this benchmark measures
Metrics
Sequential-task completion statistics.
Execution requirements
Policy adapter and CALVIN simulator.
Scope and limitations
Hold observation/action spaces, initial states and task-chain protocol fixed.
Evaluation availability
Catalog entry. Request a managed evaluation to qualify your agent interface and the benchmark’s native grading requirements.
Sources and company fit
- Original benchmark ↗Established
Companies whose products may fit
Research recommendations based on product capabilities. These companies have not necessarily run this benchmark or integrated with Blobfish.
Compatibility notes for each company
1X: Research / simulator adaptation
Figure: Research / simulator adaptation
NVIDIA: Research / simulator adaptation
Physical Intelligence: Research / simulator adaptation
Sanctuary AI: Research / simulator adaptation
Skild AI: Research / simulator adaptation