Dataset-level accuracy estimation on Spider to SynSQL 2.5M
2.8MAEFusionSQL
Evaluation Results
| Method | Links | |
|---|---|---|
| FusionSQLModel=Qwen2.5-72B2026.03 | 2.8 | |
| FusionSQL-TLModel=Qwen2.5-72B2026.03 | 3.1 | |
| FusionSQLModel=Llama-3.1-70B2026.03 | 3.2 | |
| FusionSQL-TLModel=Llama-3.1-70B2026.03 | 3.5 | |
| FusionSQLModel=DeepSeek-33B2026.03 | 3.7 | |
| FusionSQLModel=Avg.2026.03 | 3.7 | |
| FusionSQL-TLModel=DeepSeek-33B2026.03 | 4 | |
| FusionSQL-TLModel=Avg.2026.03 | 4 | |
| FusionSQLModel=XiYanSQL-14B2026.03 | 4.1 | |
| FusionSQL-TLModel=XiYanSQL-14B2026.03 | 4.4 | |
| FusionSQLModel=CSC-SQL-7B2026.03 | 4.5 | |
| FusionSQL-TLModel=CSC-SQL-7B2026.03 | 4.9 | |
| ArenaCmpModel=Qwen2.5-72B2026.03 | 8.4 | |
| ArenaCmpModel=Llama-3.1-70B2026.03 | 9.1 | |
| PseAutoEvalModel=Qwen2.5-72B2026.03 | 9.5 | |
| ArenaCmpModel=DeepSeek-33B2026.03 | 9.8 | |
| PseAutoEvalModel=Llama-3.1-70B2026.03 | 10.1 | |
| ArenaCmpModel=Avg.2026.03 | 10.1 | |
| PseAutoEvalModel=DeepSeek-33B2026.03 | 10.8 | |
| ATC-MCModel=Qwen2.5-72B2026.03 | 10.9 | |
| PseAutoEvalModel=Avg.2026.03 | 11.1 | |
| ArenaCmpModel=XiYanSQL-14B2026.03 | 11.1 | |
| ATC-MCModel=Llama-3.1-70B2026.03 | 11.7 | |
| ArenaCmpModel=CSC-SQL-7B2026.03 | 11.9 | |
| PseAutoEvalModel=XiYanSQL-14B2026.03 | 12 | |
| ATC-NEModel=Qwen2.5-72B2026.03 | 12.1 | |
| ATC-MCModel=DeepSeek-33B2026.03 | 12.3 | |
| BugJudgeModel=Qwen2.5-72B2026.03 | 12.4 | |
| ATC-MCModel=Avg.2026.03 | 12.7 | |
| ATC-NEModel=Llama-3.1-70B2026.03 | 12.9 | |
| DoC (τ=0.8)Model=Qwen2.5-72B2026.03 | 12.9 | |
| PseAutoEvalModel=CSC-SQL-7B2026.03 | 13.1 | |
| BugJudgeModel=Llama-3.1-70B2026.03 | 13.2 | |
| ATC-NEModel=DeepSeek-33B2026.03 | 13.5 | |
| DoC (τ=0.8)Model=Llama-3.1-70B2026.03 | 13.6 | |
| ATC-MCModel=XiYanSQL-14B2026.03 | 13.8 | |
| ATC-NEModel=Avg.2026.03 | 13.8 | |
| BugJudgeModel=DeepSeek-33B2026.03 | 14 | |
| DoC (τ=0.9)Model=Qwen2.5-72B2026.03 | 14.1 | |
| BugJudgeModel=Avg.2026.03 | 14.3 | |
| ATC-MCModel=CSC-SQL-7B2026.03 | 14.7 | |
| DoC (τ=0.8)Model=DeepSeek-33B2026.03 | 14.7 | |
| DoC (τ=0.9)Model=Llama-3.1-70B2026.03 | 14.8 | |
| ATC-NEModel=XiYanSQL-14B2026.03 | 14.9 | |
| DoC (τ=0.8)Model=Avg.2026.03 | 14.9 | |
| BugJudgeModel=XiYanSQL-14B2026.03 | 15.5 | |
| ATC-NEModel=CSC-SQL-7B2026.03 | 15.8 | |
| DoC (τ=0.9)Model=DeepSeek-33B2026.03 | 15.9 | |
| DoC (τ=0.8)Model=XiYanSQL-14B2026.03 | 16 | |
| DoC (τ=0.9)Model=Avg.2026.03 | 16.1 | |
| BugJudgeModel=CSC-SQL-7B2026.03 | 16.6 | |
| DoC (τ=0.8)Model=CSC-SQL-7B2026.03 | 17.2 | |
| DoC (τ=0.9)Model=XiYanSQL-14B2026.03 | 17.2 | |
| DoC (τ=0.9)Model=CSC-SQL-7B2026.03 | 18.4 |