Tool Hallucination
Arena · 50 models
| Rank | Model | Score |
|---|---|---|
| 1 | GLM-5.3-Flash | 0.008 |
| 2 | GPT-6 Astra (Max) | 0.008 |
| 3 | GLM-5.3 (Max) | 0.008 |
| 4 | DeepSeek-V4-Pro (High, 20260813) | 0.008 |
| 5 | GPT-6 Luna (Low) | 0.008 |
| 6 | GPT-6 Sol (Max) | 0.008 |
| 7 | Grok 4.6 (xHigh) | 0.008 |
| 8 | GPT-6.1 Terra (Max) | 0.008 |
| 9 | GPT-6 Luna (Medium) | 0.008 |
| 10 | Kimi K3.1 Pro (Max) | 0.008 |
| 11 | GPT-5.6 Sol (xHigh) | 0.008 |
| 12 | GPT-6.1 Sol (Max) | 0.008 |
| 13 | Kimi K3 (Max) | 0.008 |
| 14 | Grok 4.7 (xHigh) | 0.008 |
| 15 | GPT-6 Luna (Max) | 0.008 |
| 16 | GLM-5.2 (Max) | 0.008 |
| 17 | GPT-5.6 Terra (xHigh) | 0.0078 |
| 18 | Muse Spark 1.1 | 0.0076 |
| 19 | Muse Spark 1.3 (Max) | 0.0075 |
| 20 | Claude Opus 5 (Max) | 0.0075 |
| 21 | Claude Sonnet 5.5 (Max) | 0.0071 |
| 22 | Claude Fable 5.1 (Max) | 0.0068 |
| 23 | Muse Spark 1.2 (xHigh) | 0.0064 |
| 24 | Claude Opus 5.5 (Max) | 0.0054 |
| 25 | Gemini 4 Carbon (High) | 0.0053 |
| 26 | Claude Fable 5 (High) | 0.005 |
| 27 | Gemini 3.7 Flash (High) | 0.0049 |
| 28 | Claude Sonnet 5 (High) | 0.0048 |
| 29 | Gemini 4 Argon (High) | 0.0036 |
| 30 | DeepSeek-V4.1-Flash-20261017 (Max) | 0.0035 |
| 31 | Gemini 3.8 Flash (High) | 0.0034 |
| 32 | MiniMax-M3.1-Flash | 0.0014 |
| 33 | Hy4 preview | 0.0011 |
| 34 | Step 5 Preview | 0.0001 |
| 35 | Qwen3.8-27B | -0.0001 |
| 36 | Gemini 3.1 Pro Preview | -0.0005 |
| 37 | MiMo-V2.6-Pro | -0.0021 |
| 38 | Qwen3.7-Max | -0.0039 |
| 39 | Muse Wonder Full (Max) | -0.0045 |
| 40 | Qwen3.8-Max | -0.0073 |
| 41 | Beam 1.0 501B A23B | -0.0082 |
| 42 | Qwen3.8-Flash-Next | -0.0082 |
| 43 | Muse Wonder (Max) | -0.0091 |
| 44 | MiMo-V2.5-Pro | -0.0105 |
| 45 | Claude Haiku 5.5 (High) | -0.014 |
| 46 | MiniMax-M3 | -0.016 |
| 47 | Nemotron 3 Ultra | -0.0167 |
| 48 | Spark-X2.6 | -0.0172 |
| 49 | Inkling | -0.018 |
| 50 | Qwen3.7-Plus | -0.0279 |