Parallel Processing on VoxMind Agentic Core Capabilities Dataset
89.52TSVoxMind
Evaluation results
| Method | Links | ||
|---|---|---|---|
| VoxMind2026.04 | 89.52 | 61.59 | |
| Gemini-2.5-proModel Category=Closed-source models, Inference Mode=direct inference with prompt2026.04 | 88.57 | 69.52 | |
| Gemini-2.5-flashModel Category=Closed-source models, Inference Mode=direct inference with prompt2026.04 | 86.67 | 68.25 | |
| Qwen3-8B+WhisperModel Category=Open-source models, Architecture=Cascaded models, Inference Mode=direct inference with prompt2026.04 | 85.71 | 46.67 | |
| GPT-4o-audioModel Category=Closed-source models, Inference Mode=direct inference with prompt2026.04 | 84.76 | 61.9 | |
| Kimi-AudioModel Category=Open-source models, Architecture=End-to-end models, Inference Mode=direct inference with prompt2026.04 | 79.05 | 55.24 | |
| Qwen2.5-OmniModel Category=Open-source models, Architecture=End-to-end models, Inference Mode=direct inference with prompt2026.04 | 65.4 | 28.57 | |
| StepAudio2Model Category=Open-source models, Architecture=End-to-end models, Inference Mode=direct inference with prompt2026.04 | 53.33 | 33.33 |