Text2SQL on Spider (test)
88.3Exec Acc (Greedy)Omini-SQL-14B
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Omini-SQL-14BBase Model=Qwen2.5-Coder-14B-Instruct, Training Set=SynSQL(2.5M)+BIRD(9.4k)+Spider(8.7k)2026.01 | 88.3 | 88.3 | 0.0006 | |
| OminiSQL-7BBase Model=Qwen2.5-Coder-7B-Instruct, Training Set=SynSQL(2.5M)+BIRD(9.4k)+Spider(8.7k)2026.01 | 87.9 | 88.9 | 0.002 | |
| SynSQL(2.5M)Backbone=Qwen2.5-Coder-7B-Instruct2025.12 | 87.9 | 88.3 | — | |
| Spider+BIRD+DataFlow-Text2SQL-90KBackbone=Qwen2.5-Coder-7B-Instruct2025.12 | 87.5 | 88.5 | — | |
| SQL-TRAIL-14BBase Model=Qwen2.5-Coder-14B-Instruct, Training Set=SynSQL(0.8k)+Spider(1k)2026.01 | 86.8 | 88.5 | 0.1 | |
| Multi-turn AgentBase Model=Qwen2.5-Coder-14B-Instruct, Training Set=-2026.01 | 86.6 | 88 | — | |
| SQL-R1-14BBase Model=Qwen2.5-Coder-14B-Instruct, Training Set=SynSQL(5k)2026.01 | 86.1 | 88.1 | -0.65 | |
| SQL-TRAIL-7BBase Model=Qwen2.5-Coder-7B-Instruct, Training Set=SynSQL(0.8k)+Spider(1k)2026.01 | 86 | 87.6 | 1.9 | |
| DataFlow-Text2SQL-90KBackbone=Qwen2.5-Coder-7B-Instruct2025.12 | 84.8 | 86 | — | |
| DataFlow-Text2SQL-50KBackbone=Qwen2.5-Coder-7B-Instruct2025.12 | 84.6 | 85.8 | — | |
| SQL-R1-7BBase Model=Qwen2.5-Coder-7B-Instruct, Training Set=SynSQL(5k)2026.01 | 83.5 | 86.1 | 0.26 | |
| GPT-4-TurboAccess=Closed-source2025.12 | 83.4 | 84.2 | — | |
| GPT-4oAccess=Closed-source2025.12 | 83.2 | 84.9 | — | |
| GPT-4o-miniAccess=Closed-source2025.12 | 82.4 | 83.7 | — | |
| SynSQL(90K)Backbone=Qwen2.5-Coder-7B-Instruct2025.12 | 82.3 | 84.4 | — | |
| Multi-turn AgentBase Model=Qwen2.5-Coder-7B-Instruct, Training Set=-2026.01 | 82.2 | 85.6 | — | |
| Qwen2.5-Coder-7B-InstructAccess=Open-source2025.12 | 82.2 | 85.6 | — | |
| Single-passBase Model=Sonnet-3.7, Training Set=-2026.01 | 82 | 83.2 | — | |
| Multi-turn AgentBase Model=Sonnet-3.7, Training Set=-2026.01 | 81.9 | 82 | — | |
| SynSQL(50K)Backbone=Qwen2.5-Coder-7B-Instruct2025.12 | 81.8 | 84.8 | — | |
| Reasoning-SQL-14BBase Model=Qwen2.5-Coder-14B-Instruct, Training Set=BIRD-train(8k)2026.01 | 81.4 | — | -0.1 | |
| Reasoning-SQL-7BBase Model=Qwen2.5-Coder-7B-Instruct, Training Set=BIRD-train(8026)2026.01 | 78.7 | — | -0.44 | |
| Spider+BIRD+DataFlow-Text2SQL-90KBackbone=Meta-Llama-3.1-8B-Instruct2025.12 | 78.4 | 82.3 | — | |
| SynSQL(2.5M)Backbone=Meta-Llama-3.1-8B-Instruct2025.12 | 78.3 | 82.5 | — | |
| SQL-TRAIL-3BBase Model=Qwen2.5-Coder-3B-Instruct, Training Set=SynSQL(0.8k)+Spider(1k)2026.01 | 77.7 | 84.3 | 1.3 | |
| Qwen2.5-7B-InstructAccess=Open-source2025.12 | 76.8 | 82.6 | — | |
| SQL-R1-3BBase Model=Qwen2.5-Coder-3B-Instruct, Training Set=SynSQL(5k)2026.01 | 76.5 | 78.9 | 0.28 | |
| DataFlow-Text2SQL-90KBackbone=Meta-Llama-3.1-8B-Instruct2025.12 | 75.8 | 80 | — | |
| Multi-turn AgentBase Model=Qwen2.5-Coder-3B-Instruct, Training Set=-2026.01 | 75.1 | 77.2 | — | |
| DataFlow-Text2SQL-50KBackbone=Meta-Llama-3.1-8B-Instruct2025.12 | 75.1 | 80.1 | — | |
| SynSQL(90K)Backbone=Meta-Llama-3.1-8B-Instruct2025.12 | 73.4 | 78.5 | — | |
| SynSQL(50K)Backbone=Meta-Llama-3.1-8B-Instruct2025.12 | 72.7 | 78.6 | — | |
| Meta-Llama-3.1-8B-InstructAccess=Open-source2025.12 | 72.2 | 78.5 | — | |
| DeepSeek-Coder-7B-InstructAccess=Open-source2025.12 | 70.5 | 73.2 | — | |
| Granite-3.1-8B-InstructAccess=Open-source2025.12 | 69.8 | 75.3 | — | |
| OpenCoder-8B-InstructAccess=Open-source2025.12 | 68.3 | 70.1 | — | |
| Granite-8B-Code-InstructAccess=Open-source2025.12 | 64.9 | 68.6 | — |