Kikuyu Evaluation of Audio AI Models Kikuyu Evaluation of Audio AI Models By Gooey.AI, CLEAR Global and the Gates Foundation • 30 Aug 2026 • Gooey.AI/language 0 2 4 6 8 10 12 14 16 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Accuracy Latency Lower is better More accurate Gemini 3.1 Pro — Accuracy 0.72, Latency 15.11Gemini 3.1 Pro Gemini 3.1 Pro + Omni — Accuracy 0.72, Latency 11.78Gemini 3.1 Pro + Omni Gemini 3.1 Pro + Kikuyu Whisper v2 — Accuracy 0.86, Latency 9.46Gemini 3.1 Pro + Kikuyu Whisper v2Most Accurate GPT 5.6 Sol + Omni — Accuracy 0.67, Latency 8.43GPT 5.6 Sol + Omni GPT 5.6 Sol + Kikuyu Whisper v2 — Accuracy 0.82, Latency 7.68GPT 5.6 Sol + Kikuyu Whisper v2 Claude Sonnet 5 + Omni — Accuracy 0.43, Latency 7.43Claude Sonnet 5 + Omni Claude Sonnet 5 + Kikuyu Whisper v2 — Accuracy 0.63, Latency 6.29Claude Sonnet 5 + Kikuyu Whisper v2Fastest Kimi K3 + Omni — Accuracy 0.38, Latency 8.9Kimi K3 + Omni Kimi K3 + Kikuyu Whisper v2 — Accuracy 0.53, Latency 7.87Kimi K3 + Kikuyu Whisper v2 DeepSeek V4 + Omni — Accuracy 0.41, Latency 8.2DeepSeek V4 + Omni DeepSeek V4 + Kikuyu Whisper v2 — Accuracy 0.6, Latency 7.76DeepSeek V4 + Kikuyu Whisper v2 Gemini 3.7 Flash + Omni — Accuracy 0.62, Latency 8.16Gemini 3.7 Flash + Omni Gemini 3.7 Flash + Kikuyu Whisper v2 — Accuracy 0.82, Latency 7.06Gemini 3.7 Flash + Kikuyu Whisper v2 Gemini 3.1 Pro — Accuracy 0.72, Latency 15.110 Gemini 3.1 Pro + Omni — Accuracy 0.72, Latency 11.781 Gemini 3.1 Pro + Kikuyu Whisper v2 — Accuracy 0.86, Latency 9.462 GPT 5.6 Sol + Omni — Accuracy 0.67, Latency 8.433 GPT 5.6 Sol + Kikuyu Whisper v2 — Accuracy 0.82, Latency 7.684 Claude Sonnet 5 + Omni — Accuracy 0.43, Latency 7.435 Claude Sonnet 5 + Kikuyu Whisper v2 — Accuracy 0.63, Latency 6.296 Kimi K3 + Omni — Accuracy 0.38, Latency 8.97 Kimi K3 + Kikuyu Whisper v2 — Accuracy 0.53, Latency 7.878 DeepSeek V4 + Omni — Accuracy 0.41, Latency 8.29 DeepSeek V4 + Kikuyu Whisper v2 — Accuracy 0.6, Latency 7.7610 Gemini 3.7 Flash + Omni — Accuracy 0.62, Latency 8.1611 Gemini 3.7 Flash + Kikuyu Whisper v2 — Accuracy 0.82, Latency 7.0612 © 2026 Gooey.AI