Model comparison
Gemini 2.5 Pro vs Step 3
Gemini 2.5 Pro is the stronger model overall, scoring 45.0 to 40.5 on the Noometry Index.
Last verified . 17 shared benchmarks.
Google
StepFunSummary
- They share 17 benchmarks with published results for both. Gemini 2.5 Pro scores higher in 8 categories and Step 3 in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where Gemini 2.5 Pro leads 59.8 to 40.3.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 70.3% for Gemini 2.5 Pro and 62.3% for Step 3.
- Step 3 has downloadable open weights; the other is API-only.
Side by side
| Gemini 2.5 Pro | Step 3 | |
|---|---|---|
| Provider | StepFun | |
| Noometry Index | 45.0 | 40.5 |
| Released | 2025-03-25 | — |
| Weights | Proprietary | Open |
| Context window | 1.05M | — |
| Max output | 66K | — |
| Input $ / M tokens | $1.25 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 78 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 2.5 Pro leads
Gemini 2.5 Pro: 42.4 (#101), Step 3: 40.1 (#147)
| Benchmark | Gemini 2.5 Pro | Step 3 |
|---|---|---|
| LMArena Coding | 1452 | 1367 |
| SWE-bench Verified | 57.6% | — |
| SWE-bench Verified (bash only) | 53.6% | — |
| Aider Polyglot | 83.1% | — |
| LMArena WebDev | 1227 | — |
| SciCode | 42.8% | — |
| GSO | 3.9% | — |
| WeirdML | 54% | — |
| LiveBench Coding | 85.9% | — |
| CadEval | 64% | — |
| ALE-Bench | 785.52 | — |
| AlgoTune | 1.51 | — |
Agentic & Tool Use Not comparable
Gemini 2.5 Pro: 29.2 (#88), Step 3: —
| Benchmark | Gemini 2.5 Pro | Step 3 |
|---|---|---|
| Terminal-Bench | 32.6% | — |
| GDPval | 23.3% | — |
| Remote Labor Index | 0.8% | — |
| TheAgentCompany | 30.3% | — |
| τ²-bench Banking | 13.7% | — |
| DeepResearch Bench | 42.8% | — |
| BALROG | 43.3% | — |
| LMArena Search | 1142 | — |
| METR Time Horizons | 55.4% | — |
| Vending-Bench 2 | 573.64 | — |
Reasoning Too close to call
Gemini 2.5 Pro: 28.8 (#99), Step 3: 28.4 (#105)
| Benchmark | Gemini 2.5 Pro | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 70.3% | 62.3% |
| LMArena Hard Prompts | 1455 | 1355 |
| ARC-AGI-2 | 4.9% | — |
| SimpleBench | 62.4% | — |
| ARC-AGI-1 | 41% | — |
| CritPt | 2% | — |
| Chess Puzzles | 20% | — |
| EnigmaEval | 5.6% | — |
| LiveBench Reasoning | 89.8% | — |
| DTBench | 82.4% | — |
| LiveBench Data Analysis | 79.9% | — |
| LMCA | 34.8% | — |
| Epoch Capabilities Index | 145.32 | — |
| ForecastBench | 61.3 | — |
| LiveBench | 82.3% | — |
Math Step 3 leads
Gemini 2.5 Pro: 32.5 (#213), Step 3: 37.6 (#148)
| Benchmark | Gemini 2.5 Pro | Step 3 |
|---|---|---|
| LMArena Math | 1450 | 1366 |
| FrontierMath (Tiers 1-3) | 24.6% | — |
| FrontierMath Tier 4 | 0% | — |
| OTIS Mock AIME 2024-2025 | 84.7% | — |
| Omni-MATH | 41.6% | — |
| LiveBench Math | 90.2% | — |
| MATH Level 5 | 95.9% | — |
| FrontierMath (Feb 2025 set) | 14.1% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Gemini 2.5 Pro leads
Gemini 2.5 Pro: 56.0 (#46), Step 3: 36.8 (#164)
| Benchmark | Gemini 2.5 Pro | Step 3 |
|---|---|---|
| LMArena Expert | 1452 | 1333 |
| GPQA Diamond | 85.3% | — |
| Humanity's Last Exam | 21.6% | — |
| MMLU-Pro | 86.3% | — |
| Confabulations | 10.6% | — |
| Vectara Hallucination Rate | 7% | — |
| GPQA (HELM) | 74.9% | — |
Multimodal Gemini 2.5 Pro leads
Gemini 2.5 Pro: 45.2 (#18), Step 3: 35.5 (#86)
| Benchmark | Gemini 2.5 Pro | Step 3 |
|---|---|---|
| LMArena Vision | 1263 | 1177 |
| GeoBench | 86% | — |
| VPCT | 48% | — |
| LMArena Document | 1421 | — |
| SpatialViz-Bench | 44.7% | — |
Multilingual Gemini 2.5 Pro leads
Gemini 2.5 Pro: 55.3 (#31), Step 3: 46.3 (#159)
| Benchmark | Gemini 2.5 Pro | Step 3 |
|---|---|---|
| LMArena Non-English | 1451 | 1327 |
| LMArena Chinese | 1507 | 1397 |
| LMArena German | 1487 | 1371 |
| LMArena Korean | 1434 | 1269 |
| LMArena Russian | 1461 | 1331 |
| LMArena Spanish | 1473 | 1371 |
| LMArena French | 1472 | — |
| LMArena Japanese | 1461 | — |
Instruction Following Gemini 2.5 Pro leads
Gemini 2.5 Pro: 75.0 (#75), Step 3: 70.4 (#164)
| Benchmark | Gemini 2.5 Pro | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1437 | 1332 |
| LiveBench Instruction Following | 80.6% | — |
| IFEval | 84% | — |
Long Context Gemini 2.5 Pro leads
Gemini 2.5 Pro: 59.8 (#5), Step 3: 40.3 (#157)
| Benchmark | Gemini 2.5 Pro | Step 3 |
|---|---|---|
| LMArena Longer Query | 1449 | 1326 |
| Fiction.LiveBench | 91.7% | — |
Writing & Preference Gemini 2.5 Pro leads
Gemini 2.5 Pro: 63.7 (#62), Step 3: 54.3 (#151)
| Benchmark | Gemini 2.5 Pro | Step 3 |
|---|---|---|
| LMArena Text | 1458 | 1350 |
| LMArena Creative Writing | 1454 | 1321 |
| LMArena Multi-Turn | 1453 | 1341 |
| Short-Story Creative Writing | 83.8% | — |
| EQ-Bench Creative Writing | 1421 | — |
| WildBench | 85.7% | — |
| LiveBench Language | 67.8% | — |
Frequently asked questions
Is Gemini 2.5 Pro better than Step 3?
Gemini 2.5 Pro is the stronger model overall, scoring 45.0 to 40.5 on the Noometry Index.
Is Gemini 2.5 Pro or Step 3 better for coding?
Gemini 2.5 Pro scores higher on coding benchmarks: 42.4 versus 40.1 in the Noometry coding category.
How many benchmarks do Gemini 2.5 Pro and Step 3 share?
17 benchmarks have published results for both models. Gemini 2.5 Pro has 78 scored results on Noometry and Step 3 has 17.