Gemini 3.5 Flash wins more areas, narrowly.
Scores updated · 19 tests both models report · How we compare
Where each one wins
Tests won in each of the four areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
Reasoning rests on a single test.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Gemini 3.5 Flash pulls ahead
- Long professional tasks in banking, consulting and lawAA ApexAgents+17.9points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+11.1points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+10.8points ahead
Every test, side by side
All 19 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- Terminal-Bench 2.1Gemini 3.5 Flash by 11.178.767.6+11.1
- SciCodeSeed2.1 by 3.953.957.8+3.9
AgentsGemini 3.5 Flash
- GDPValSeed2.1 by 47.635.182.7+47.6
- AA ApexAgentsGemini 3.5 Flash by 17.947.129.2+17.9
- MCP AtlasGemini 3.5 Flash by 3.383.680.3+3.3
Images and chartsEven
Other results11 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- TOMATOGemini 3.5 Flash by 15.171.956.8+15.1
- DeepSWEGemini 3.5 Flash by 143723+14
- OVBenchSeed2.1 by 13.256.569.7+13.2
- ToolathlonGemini 3.5 Flash by 7.456.549.1+7.4
- VideoSimpleQAGemini 3.5 Flash by 4.67671.4+4.6
- MotionBenchSeed2.1 by 4.270.674.8+4.2
- LongVideoBenchSeed2.1 by 377.680.6+3
- MinervaGemini 3.5 Flash by 2.768.665.9+2.7
- TVBenchtie76.477.2tie
- LVBenchtie76.376.8tie
- VideoHolmestie67.167.6tie
Questions people ask
Which is better, Gemini 3.5 Flash or Seed2.1?
Gemini 3.5 Flash wins two of the four areas where both have results: agents and reasoning. Seed2.1 wins none. They are level on coding and images and charts.
Which is better for coding?
Neither. They win 1 coding test each of the 2 both models report.
How do you compare the two?
We use the 19 benchmark tests both models have published scores on. The verdict counts the 8 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 11 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.