Seed2.1 Model Card: Agentic Intelligence for Productivity
Open paper
Benchmark
Model
Rank
Results
code-generation-on-swe-bench-pro
Seed2.1 Pro
–
Resolved: 57.5
code-generation-on-terminal-bench-2-1
Seed2.1 Pro
–
Accuracy: 71.0
Rank counts only results with a code link.