Gemini 3.1 Pro Preview coding benchmark

How Gemini 3.1 Pro Preview performed

Gemini 3.1 Pro Preview by Google took the PomBench AI coding benchmark: every model gets the same prompt and writes self-contained browser code for a twelve-second scene: a Pomeranian runs across a backyard, jumps a puddle, enters its doghouse, retrieves a bone and turns toward the viewer. The video above is an independent offline browser recording of the code Gemini 3.1 Pro Preview returned, captured at 960 × 540 and 30 fps, with no organizer edits to the code.

In this Gemini 3.1 Pro Preview coding test the model produced 1,920 output tokens, including provider-reported reasoning (rank 42 of 46 by output tokens), and the known API cost was $0.0273 (rank 17 of 45, cheapest first). Is Gemini 3.1 Pro Preview good at coding? One task cannot answer that on its own, so this LLM coding benchmark lets you watch the result instead of reading a score: check the gait, the puddle jump and reflection, the doorway occlusion and whether the bone stays in the dog's mouth.

Then compare Gemini 3.1 Pro Preview vs Gemini 3.5 Flash-Lite and Gemini 3.8 Flash below, or open the AI model comparison to play up to four models on one synchronized timeline.