Agent Capability Evaluation on ACEBench en
71.4Normal ScoreQwen3-8B
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Qwen3-8BBackbone=Qwen3-8B-Base, Thinking Mode=true2026.06 | 71.4 | 75.3 | 34.2 | |
| Qwen3-4BBackbone=Qwen3-4B-B., Thinking Mode=true2026.06 | 70.6 | 68.7 | 20 | |
| GAISBackbone=Qwen3-8B-Base, Thinking Mode=true2026.06 | 67.7 | 61.3 | 20.8 | |
| NemotronBackbone=Qwen3-14B-B., Thinking Mode=true2026.06 | 67.7 | 61.7 | 7.5 | |
| GAISBackbone=Qwen3-14B-B., Thinking Mode=true2026.06 | 67.7 | 77.3 | 23.3 | |
| GAISBackbone Scale=8B, Fine-tuning Dataset=GAIS2026.06 | 67.2 | 27.3 | 11.7 | |
| Qwen3-14BBackbone=Qwen3-14B-B., Thinking Mode=true2026.06 | 66.9 | 84 | 41.6 | |
| Qwen3-8BBackbone=Qwen3-8B-Base, Thinking Mode=false2026.06 | 66.5 | 28 | 11.7 | |
| Qwen3-8BBackbone Scale=8B, Fine-tuning Dataset=Official instruction-tuned2026.06 | 66.5 | 28 | 11.7 | |
| NemotronBackbone=Qwen3-8B-Base, Thinking Mode=true2026.06 | 65.8 | 40.7 | 7.5 | |
| Qwen3-14BBackbone=Qwen3-14B-B., Thinking Mode=false2026.06 | 65.8 | 53.3 | 11.7 | |
| GAISBackbone Scale=14B, Fine-tuning Dataset=GAIS2026.06 | 65.8 | 58.7 | 14.2 | |
| Qwen3-14BBackbone Scale=14B, Fine-tuning Dataset=Official instruction-tuned2026.06 | 65.2 | 53.3 | 11.7 | |
| NemotronBackbone Scale=14B, Fine-tuning Dataset=Nemotron2026.06 | 64.6 | 42.7 | 14.2 | |
| GAISBackbone=Qwen3-14B-B., Thinking Mode=false2026.06 | 62.7 | 48.7 | 5.8 | |
| NemotronBackbone=Qwen3-14B-B., Thinking Mode=false2026.06 | 61.9 | 46.3 | 0 | |
| Qwen3-4BBackbone=Qwen3-4B-B., Thinking Mode=false2026.06 | 61.8 | 29.3 | 1.7 | |
| Qwen3-4BBackbone Scale=4B, Fine-tuning Dataset=Official instruction-tuned2026.06 | 61.8 | 29.3 | 1.7 | |
| NemotronBackbone Scale=4B, Fine-tuning Dataset=Nemotron2026.06 | 59.4 | 32 | 1.7 | |
| GAISBackbone Scale=4B, Fine-tuning Dataset=GAIS2026.06 | 59.4 | 36.7 | 4.2 | |
| GAISBackbone=Qwen3-4B-B., Thinking Mode=true2026.06 | 58.2 | 59.3 | 7.5 | |
| GAISBackbone=Qwen3-8B-Base, Thinking Mode=false2026.06 | 53.2 | 24 | 7.5 | |
| NemotronBackbone=Qwen3-8B-Base, Thinking Mode=false2026.06 | 52.2 | 32.7 | 0 | |
| NemotronBackbone Scale=8B, Fine-tuning Dataset=Nemotron2026.06 | 52.2 | 32.7 | 0 | |
| GAISBackbone=Qwen3-4B-B., Thinking Mode=false2026.06 | 49.1 | 37.3 | 0 | |
| Llama-3.1-8B-Ins.Backbone=Llama, Thinking Mode=false2026.06 | 46.6 | 21 | 5.3 | |
| NemotronBackbone=Qwen3-4B-B., Thinking Mode=true2026.06 | 42.6 | 54 | 3.7 | |
| GAISBackbone=Llama, Thinking Mode=false2026.06 | 42.5 | 19.8 | 7 | |
| NemotronBackbone=Llama, Thinking Mode=false2026.06 | 37.8 | 23.5 | 0 | |
| ToolACEBackbone=Qwen3-8B-Base, Thinking Mode=false2026.06 | 28.3 | 0 | 4.2 | |
| ToolACEBackbone Scale=8B, Fine-tuning Dataset=ToolACE2026.06 | 28.3 | 0 | 4.2 | |
| NemotronBackbone=Qwen3-4B-B., Thinking Mode=false2026.06 | 7 | 26 | 0 | |
| ToolLLMBackbone=Qwen3-8B-Base, Thinking Mode=false2026.06 | 0 | 0 | 0 | |
| ToolLLMBackbone Scale=8B, Fine-tuning Dataset=ToolLLM2026.06 | 0 | 0 | 0 |