GeminivsGrok

Comparing Google and xAI. Average Podium Score across top models: Gemini (72.8) vs Grok (73.6). Flagship showdown: Gemini 3.1 Pro Preview (74.0) vs Grok 4.20 Beta1 (74.0).

Gemini (Google)

Native multimodal models with massive context windows from Google DeepMind.

Avg Score: 72.8
Licensing: Proprietary API

Grok (xAI)

Real-time knowledge and frontier reasoning models from xAI.

Top Model: Grok 4.20 Beta1
Avg Score: 73.6
Licensing: Proprietary API

Top Models Showdown

RankModelLabScoreCodingReasoningSpeedPrice / 1M
#1Gemini 3.1 Pro PreviewGoogle74.050 t/s$5/M
#2Gemini 3.5 Flash HighGoogle74.050 t/s$5/M
#3Gemini 3.5 Flash MediumGoogle74.050 t/s$5/M
#4Grok 4.20 Beta1xAI74.050 t/s$15/M
#5Grok 4.20 Beta 0309 ReasoningxAI74.050 t/s$15/M
#6Grok 4.20 Multi Agent Beta 0309xAI74.050 t/s$15/M
#7Grok 4.1 ThinkingxAI73.050 t/s$15/M
#8Grok 4.1xAI73.050 t/s$15/M
#9Gemini 3.1 Flash Lite PreviewGoogle72.050 t/s$5/M
#10Gemini 2.5 Flash Preview 09 2025Google70.050 t/s$5/M
0 / 4 Models Selected