Multi-turn RAG Generation on RAG Evaluation (dev)
67RB_llm5ting
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| 5ting2026.06 | 67 | 77 | 38 | 55 | |
| GPT-4o2026.06 | 66 | 65 | 38 | 53 | |
| Qwen 2.5Size=72B2026.06 | 65 | 64 | 37 | 52 | |
| GPT-4o-mini2026.06 | 64 | 64 | 37 | 52 | |
| Llama 3.1 InstructSize=405B2026.06 | 63 | 65 | 39 | 53 | |
| Mixtral 8x22BSize=141B2026.06 | 61 | 56 | 35 | 48 | |
| Llama 3.1 InstructSize=70B2026.06 | 59 | 64 | 39 | 52 | |
| Command-R+Size=104B2026.06 | 59 | 66 | 38 | 51 | |
| Llama 3.1 InstructSize=8B2026.06 | 54 | 53 | 34 | 45 |