# Gemini 3.1 Pro review: 8.3/10

URL: https://ongchummo.vn/en/review-ai-tool/gemini-3-1-pro

Category: Multimodal LLM · Pricing: Free plan available
Tool website: https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-pro/

## Verdict

Google's highest Pro version, very strong at vision but now lagging in overall score.

## Scores by task (out of 10)

- Video: 9.2
- Image: 9.4
- 3D: 8.2
- Chat: 8.6
- Code: 8.2
- Web: 8.4

## Strengths

- ARC-AGI-1 at 98%, GPQA Diamond at 94.1%
- Image and video handling is among the best
- Tiered pricing by size, $2 to $4 input and $12 to $18 output

## Weaknesses

- AA Index 48, has fallen behind the leading group
- Tiered pricing makes costs hard to estimate
- Gemini 3.5 Pro is months late, so there is no successor yet

## Full review

### Ranking source
Updated August 9, 2026 from the Artificial Analysis Intelligence Index (v4.1 scale), cross-checked against each provider's release notes and official pages. Scores on the v4.1 scale can't be compared directly with earlier rankings.

### When to use it
Use it when the problem is heavy on vision: reading scanned documents, analyzing video, checking images against rules.

### Notes
AA Intelligence Index: 48; context 1M; price $2 to $4 input and $12 to $18 output depending on the size tier. For pure text and code work, Terra or Sonnet 5 score higher at a similar price.
