Model comparison
Gemini 2.5 Flash vs Gemini 3 Pro
Gemini 3 Pro is the stronger model overall, scoring 54.7 to 39.0 on the Noometry Index.
Last verified . 46 shared benchmarks.
GoogleSummary
- They share 46 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 1 category and Gemini 3 Pro in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Pro leads 52.5 to 18.2.
- The biggest single-benchmark swing is Terminal-Bench: 17.1% for Gemini 2.5 Flash and 69.4% for Gemini 3 Pro.
Side by side
| Gemini 2.5 Flash | Gemini 3 Pro | |
|---|---|---|
| Provider | ||
| Noometry Index | 39.0 | 54.7 |
| Released | 2025-04-17 | 2025-11-18 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | — |
| Max output | 66K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $2.50 | — |
| Results tracked | 54 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3 Pro leads
Gemini 2.5 Flash: 35.5 (#210), Gemini 3 Pro: 52.1 (#33)
| Benchmark | Gemini 2.5 Flash | Gemini 3 Pro |
|---|---|---|
| SWE-bench Verified (bash only) | 28.7% | 74.2% |
| WeirdML | 41.9% | 69.9% |
| LMArena Coding | 1424 | 1482 |
| ALE-Bench | 661.88 | 1,177 |
| SWE-bench Verified | — | 72.9% |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1439 |
| SWE-bench Multilingual | — | 68.7% |
| GSO | — | 18.6% |
| AlgoTune | — | 1.83 |
Agentic & Tool Use Gemini 3 Pro leads
Gemini 2.5 Flash: 30.6 (#67), Gemini 3 Pro: 40.5 (#21)
| Benchmark | Gemini 2.5 Flash | Gemini 3 Pro |
|---|---|---|
| Terminal-Bench | 17.1% | 69.4% |
| Berkeley Function Calling Leaderboard | 56.2% | 72.5% |
| BALROG | 33.5% | 58.1% |
| Vending-Bench 2 | 548.84 | 5,478 |
| GDPval | — | 40.3% |
| Remote Labor Index | — | 1.3% |
| TheAgentCompany | 41.1% | — |
| τ²-bench Airline | — | 80.5% |
| τ²-bench Banking | — | 18% |
| τ²-bench Retail | — | 75.9% |
| τ²-bench Telecom | — | 91% |
| DeepResearch Bench | — | 46.3% |
| LMArena Search | — | 1207 |
| METR Time Horizons | — | 71% |
Reasoning Gemini 3 Pro leads
Gemini 2.5 Flash: 18.2 (#271), Gemini 3 Pro: 52.5 (#28)
| Benchmark | Gemini 2.5 Flash | Gemini 3 Pro |
|---|---|---|
| ARC-AGI-2 | 2.5% | 31.1% |
| SimpleBench | 41.2% | 76.4% |
| Kagi LLM Benchmark | 56.8% | 80.1% |
| ARC-AGI-1 | 33.3% | 75% |
| CritPt | 1.1% | 6.9% |
| EnigmaEval | 2.7% | 18.2% |
| LMArena Hard Prompts | 1422 | 1481 |
| Epoch Capabilities Index | 143.04 | 152.96 |
| ForecastBench | 60.6 | 61.2 |
| NYT Connections (extended) | — | 94.4% |
| Chess Puzzles | — | 31% |
| DTBench | 76.5% | — |
| LMCA | 27.5% | — |
Math Gemini 3 Pro leads
Gemini 2.5 Flash: 38.1 (#87), Gemini 3 Pro: 48.8 (#50)
| Benchmark | Gemini 2.5 Flash | Gemini 3 Pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 91.4% |
| Omni-MATH | 38.5% | 55.5% |
| LMArena Math | 1415 | 1475 |
| FrontierMath (Feb 2025 set) | 4.8% | 37.6% |
| FrontierMath Tier 4 (v1) | 4.2% | 18.8% |
| MathArena Final-Answer Competitions | — | 67% |
| ProofBench | — | 20% |
Knowledge Gemini 3 Pro leads
Gemini 2.5 Flash: 36.1 (#162), Gemini 3 Pro: 64.3 (#15)
| Benchmark | Gemini 2.5 Flash | Gemini 3 Pro |
|---|---|---|
| Humanity's Last Exam | 12.1% | 37.5% |
| MMLU-Pro | 63.9% | 90.3% |
| Vectara Hallucination Rate | 7.8% | 13.6% |
| GPQA (HELM) | 39% | 80.3% |
| LMArena Expert | 1426 | 1475 |
| GPQA Diamond | — | 92.6% |
| Confabulations | 16.8% | — |
Multimodal Gemini 3 Pro leads
Gemini 2.5 Flash: 41.2 (#31), Gemini 3 Pro: 56.9 (#1)
| Benchmark | Gemini 2.5 Flash | Gemini 3 Pro |
|---|---|---|
| LMArena Vision | 1254 | 1305 |
| GeoBench | 76% | 84% |
| VPCT | 46.2% | 91% |
| LMArena Document | — | 1434 |
| SpatialViz-Bench | 36.9% | — |
Multilingual Gemini 3 Pro leads
Gemini 2.5 Flash: 52.5 (#83), Gemini 3 Pro: 57.1 (#14)
| Benchmark | Gemini 2.5 Flash | Gemini 3 Pro |
|---|---|---|
| LMArena Non-English | 1410 | 1474 |
| LMArena Chinese | 1450 | 1522 |
| LMArena French | 1433 | 1492 |
| LMArena German | 1419 | 1517 |
| LMArena Japanese | 1403 | 1513 |
| LMArena Korean | 1385 | 1449 |
| LMArena Russian | 1415 | 1493 |
| LMArena Spanish | 1421 | 1468 |
Instruction Following Too close to call
Gemini 2.5 Flash: 75.9 (#50), Gemini 3 Pro: 76.4 (#42)
| Benchmark | Gemini 2.5 Flash | Gemini 3 Pro |
|---|---|---|
| IFEval | 89.8% | 87.7% |
| LMArena Instruction Following | 1405 | 1458 |
Long Context Gemini 2.5 Flash leads
Gemini 2.5 Flash: 47.5 (#16), Gemini 3 Pro: 44.0 (#75)
| Benchmark | Gemini 2.5 Flash | Gemini 3 Pro |
|---|---|---|
| LMArena Longer Query | 1419 | 1472 |
| Fiction.LiveBench | 77.8% | — |
| CL-bench | — | 15.8% |
Writing & Preference Gemini 3 Pro leads
Gemini 2.5 Flash: 54.0 (#153), Gemini 3 Pro: 66.6 (#34)
| Benchmark | Gemini 2.5 Flash | Gemini 3 Pro |
|---|---|---|
| LMArena Text | 1417 | 1479 |
| LMArena Creative Writing | 1400 | 1482 |
| EQ-Bench Creative Writing | 1137 | 1525 |
| WildBench | 81.7% | 85.9% |
| LMArena Multi-Turn | 1408 | 1484 |
| Short-Story Creative Writing | 76.5% | — |
Frequently asked questions
Is Gemini 2.5 Flash better than Gemini 3 Pro?
Gemini 3 Pro is the stronger model overall, scoring 54.7 to 39.0 on the Noometry Index.
Is Gemini 2.5 Flash or Gemini 3 Pro better for coding?
Gemini 3 Pro scores higher on coding benchmarks: 52.1 versus 35.5 in the Noometry coding category.
How many benchmarks do Gemini 2.5 Flash and Gemini 3 Pro share?
46 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Gemini 3 Pro has 67.