Solar System
loads medium
Tech. quality81
UX76
Time1 min 55 s
Cost$0.23
Compare →
Gemini 3.1 Pro
Loads without errors5/5#1 of 26
Tech. quality90/100#19 of 26
UX86/100#11 of 26
Total cost$1.12
Average time1 min 57 s#6 of 26
Total tokens95,232
API price: $2 / $12 per 1M input / output tokens · as of 17/10/2026 source ↗
What is Gemini 3.1 Pro good for?
The third-generation Gemini Pro model, optimised for software engineering and agentic workflows that need precise tool use.
Description based on the maker · source ↗
- Maker
- Released
- 19/02/2026
- Context window
- 1.05M tokens
- Model weights
- closed, the model is available through the API only
- Reasoning levels
- low medium high
How did Gemini 3.1 Pro do in the AI Lab tests?
- 5 of 5 outputs load without errors.
- Its best technical result was on the “New York & Central Park map” prompt (96/100).
- It implemented the chess rules flawlessly – passing the objective perft test on all 6 positions.
- Average time per prompt 1 min 57 s#6 of 26
- Average cost per prompt $0.22#16 of 26
- Technical quality 90/100#19 of 26
- Automatic UX rating 86/100#11 of 26
Outputs
All outputs of Gemini 3.1 Pro on the 5 prompts every other model got as well.
GTA-style game
loads medium
Tech. quality89
UX100
Time2 min 06 s
Cost$0.24
Compare →
Monitor test
loads medium
Tech. quality89
UX100
Time1 min 09 s
Cost$0.12
Compare →
New York & Central Park map
loads medium
Tech. quality96
UX81
Time1 min 55 s
Cost$0.21
Compare →
Chess with an engine
loads medium
Tech. quality95
UX72
Time2 min 38 s
Cost$0.31
Compare →
Cost = tokens priced with the API price list valid on the date shown next to the cost (prices change). ≈ = estimate (only the total token count is known), no ≈ = exact from input / output tokens. “Loads” = the page opens without errors; technical quality also checks behaviour during use – judge the quality in the viewer.