Result Feedback on VoxMind Agentic Core Capabilities
4.22FCGPT-4o-audio
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-4o-audioModel Category=Closed-source models, Inference Mode=direct inference with prompt2026.04 | 4.22 | |
| Gemini-2.5-proModel Category=Closed-source models, Inference Mode=direct inference with prompt2026.04 | 4.16 | |
| Gemini-2.5-flashModel Category=Closed-source models, Inference Mode=direct inference with prompt2026.04 | 4.1 | |
| Qwen3-8B+WhisperModel Category=Open-source models, Architecture=Cascaded models, Inference Mode=direct inference with prompt2026.04 | 4.05 | |
| VoxMind2026.04 | 3.94 | |
| Kimi-AudioModel Category=Open-source models, Architecture=End-to-end models, Inference Mode=direct inference with prompt2026.04 | 3.62 | |
| Qwen2.5-OmniModel Category=Open-source models, Architecture=End-to-end models, Inference Mode=direct inference with prompt2026.04 | 2.82 | |
| StepAudio2Model Category=Open-source models, Architecture=End-to-end models, Inference Mode=direct inference with prompt2026.04 | 1.91 |