Text Generation on AbGen
4.91ImportanceROMA
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| ROMABase Model=DeepSeek-V3.2-Exp, Source Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.91 | 4.64 | 4.62 | 4.93 | |
| GPT-4.1Source Type=Closed-source, Judge Model=GPT-4.1-mini2026.02 | 4.78 | 4.72 | 4.28 | 4.86 | |
| GPT-4oSource Type=Closed-source, Judge Model=GPT-4.1-mini2026.02 | 4.76 | 4.58 | 4.11 | 4.76 | |
| DeepSeek-V3.2-ExpSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.74 | 4.75 | 4.17 | 4.79 | |
| deepseek-reasonerSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.73 | 4.69 | 4.32 | 4.9 | |
| Qwen3-32B-AWQSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.72 | 4.43 | 4.19 | 4.73 | |
| Llama-4-Scout-17B-16E-InstructSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.69 | 4.1 | 4.01 | 4.38 | |
| Qwen3-235B-A22B-fp8-tputSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.69 | 4.58 | 4.18 | 4.84 | |
| Qwen3-8BSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.68 | 4.42 | 4.09 | 4.75 | |
| deepseek-chatSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.68 | 4.54 | 4.1 | 4.77 | |
| Meta-Llama-3.1-70B-Instruct-FP8Source Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.67 | 4.15 | 4.04 | 4.4 | |
| Mistral-Small-3.1-24B-Instruct-2503Source Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.67 | 4.4 | 4.1 | 4.66 | |
| Qwen2.5-32B-InstructSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.67 | 4.17 | 4.07 | 4.61 | |
| o4-miniSource Type=Closed-source, Judge Model=GPT-4.1-mini2026.02 | 4.66 | 4.59 | 4.27 | 4.78 | |
| Gemini 2.5 Flash Preview-05-20Source Type=Closed-source, Judge Model=GPT-4.1-mini2026.02 | 4.65 | 4.64 | 4.15 | 4.66 | |
| DeepSeek-R1-0528-Qwen3-8BSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.59 | 4.5 | 4.1 | 4.61 | |
| gemma-3-27b-itSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.58 | 4.53 | 4.2 | 4.59 | |
| phi-4Source Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.55 | 4.42 | 4.09 | 4.61 | |
| Llama-3.3-70B-Instruct-FP8-dynamicSource Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.51 | 4.14 | 4.03 | 4.34 | |
| Llama-4-Maverick-17B-128E-Instruct-FP8Source Type=Open-source, Judge Model=GPT-4.1-mini2026.02 | 4.5 | 4.42 | 4.03 | 4.56 |