Solar System
broken medium
Tech. quality81
UX72
Time–
Cost$0.12
Compare →
Gemini 3.8 Flash
Loads without errors4/5#12 of 26
Tech. quality87/100#23 of 26
UX85/100#13 of 26
Total cost$0.57
Average time3 min 21 s#13 of 26
Total tokens152,540
API price: $0.75 / $3.75 per 1M input / output tokens · as of 16/10/2026 source ↗
What is Gemini 3.8 Flash good for?
Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents and complex enterprise workflows.
Description based on the maker · source ↗
- Maker
- Released
- 02/09/2026
- Model weights
- closed, the model is available through the API only
- Reasoning levels
- low medium high
How did Gemini 3.8 Flash do in the AI Lab tests?
- 4 of 5 outputs load without errors.
- Its best technical result was on the “GTA-style game” prompt (98/100).
- Did not work (original output): Solar System.
- In the objective perft chess test it passed 0 of 6 positions.
- Average time per prompt 3 min 21 s#13 of 26
- Average cost per prompt $0.11#13 of 26
- Technical quality 87/100#23 of 26
- Automatic UX rating 85/100#13 of 26
Outputs
All outputs of Gemini 3.8 Flash on the 5 prompts every other model got as well.
Monitor test
loads medium
Tech. quality96
UX83
Time2 min 31 s
Cost$0.10
Compare →
GTA-style game
loads medium
Tech. quality98
UX100
Time3 min 10 s
Cost$0.10
Compare →
New York & Central Park map
loads medium
Tech. quality81
UX73
Time4 min 03 s
Cost$0.12
Compare →
Chess with an engine
loads medium
Tech. quality78
UX96
Time3 min 38 s
Cost$0.12
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.