Gemini 3.1 Flash Image
Gemini 3.1 Flash Image ranks first on 2 of the 15 benchmarks with five or more models scored.
Gemini 3.1 Flash Image
Closed weights
- RISEBench++ 50.8% #3 /22
- SolveEdit 55.5% #3 /9
- PaintBench 17.1% #1 /8
Where has it been number one?
Gemini 3.1 Flash Image led 1 benchmark at some point, for a median of 204 days. It still tops 1 that nobody has tested since.
Source: Official leaderboards
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SnakeBench | 16 | #177 of 201 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LLM Agronomy Benchmark (Farmers Business Network edition) | 90.8% | #20 of 94 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| StudyArena Image | 1,262 | #3 of 29 | Official leaderboard | |
| UltraText Bench | 97.0 | #2 of 17 | Official leaderboard | |
| PaintBench | 17.1% | #1 of 8 | Official leaderboard | |
| LLM Agronomy Benchmark Image Identification (four choices) | 87.3% | #5 of 37 | Official leaderboard | |
| RISEBench++ | 50.8% | #3 of 22 | Official leaderboard | |
| Qwen-Image-Bench | 59.8 | #2 of 13 | Official leaderboard | |
| ConceptEdit-Bench | 66.2 | #1 of 6 | Official leaderboard | |
| DesignArena Image | 1,244 | #12 of 52 | Official leaderboard | |
| T2LSC-Bench | 99% | #1 of 4 | Official leaderboard | |
| DesignArena Logo | 1,248 | #12 of 47 | Official leaderboard | |
| MDMR-Bench | 81.2% | #2 of 7 | Official leaderboard | |
| DesignArena Graphic Design | 1,249 | #14 of 46 | Official leaderboard | |
| SolveEdit | 55.5% | #3 of 9 | Official leaderboard | |
| Tricky Edits of Collected and Curated Images | 19.5% | #1 of 3 | Official leaderboard | |
| MoI-Bench | 0.95 | #3 of 5 | Official leaderboard | |
| GenPuzzle | 32.4 | #2 of 3 | Official leaderboard | |
| LongBanana | 74.6 | #1 of 1 | Official leaderboard |
What does it cost to run?
On the one leaderboard that publishes a price, its best run costs about 1/9 of the median model's. There, another model comes close to its score for less.
| Benchmark | Score | Price | From cheapest to priciest | Against the median | Close to its score for less Close for less |
|---|---|---|---|---|---|
| SnakeBench | 16 | $0.0059 0.1× the median | 0.1× | |
Inside agents
Boards that list Gemini 3.1 Flash Image inside an agent or product, such as OpenClaw + RPR + Gemini 3.1 Flash Image. Those scores belong to the agent, so we don't count them as the model's own. Its own score on the same board is beside each one.
| Benchmark | Entry | Agent's score | Its own score |
|---|---|---|---|
| LongBanana | OpenClaw + RPR + Gemini 3.1 Flash Image | 78.4 | 74.6 |
| LongBanana | OpenCode + RPR + Gemini 3.1 Flash Image | 77.8 | 74.6 |
| LongBanana | OpenCode + Gemini 3.1 Flash Image | 77.3 | 74.6 |
| LongBanana | OpenClaw + Gemini 3.1 Flash Image | 76.1 | 74.6 |
| LongBanana | Hermes-Agent + RPR + Gemini 3.1 Flash Image | 75.6 | 74.6 |
| LongBanana | Hermes-Agent + Gemini 3.1 Flash Image | 73.9 | 74.6 |
Who measured it?
Independent runs cover all 19 of its benchmarks.
- An independent run only 19
Sources: Official leaderboards, Epoch AI (CC BY 4.0) and labs’ model cards