Model comparison
Gemini 1.5 Pro (May 2024) vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 32.1 on the Noometry Index.
Last verified . 16 shared benchmarks.
Google
StepFunSummary
- They share 16 benchmarks with published results for both. Gemini 1.5 Pro (May 2024) scores higher in 1 category and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 3 leads 28.4 to 12.3.
- Step 3 has downloadable open weights; the other is API-only.
Side by side
| Gemini 1.5 Pro (May 2024) | Step 3 | |
|---|---|---|
| Provider | StepFun | |
| Noometry Index | 32.1 | 40.5 |
| Released | 2024-02-15 | — |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 45 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Gemini 1.5 Pro (May 2024): 34.2 (#241), Step 3: 40.1 (#147)
| Benchmark | Gemini 1.5 Pro (May 2024) | Step 3 |
|---|---|---|
| LMArena Coding | 1294 | 1367 |
| WeirdML | 22.2% | — |
| BigCodeBench Instruct | 43.8% | — |
| BigCodeBench Complete | 57.5% | — |
| CadEval | 34% | — |
| HumanEval+ | 79.3% | — |
| MBPP+ | 74.6% | — |
Agentic & Tool Use Not comparable
Gemini 1.5 Pro (May 2024): 17.9 (#145), Step 3: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Step 3 |
|---|---|---|
| TheAgentCompany | 3.4% | — |
| Cybench | 7.5% | — |
| BALROG | 21% | — |
Reasoning Step 3 leads
Gemini 1.5 Pro (May 2024): 12.3 (#338), Step 3: 28.4 (#105)
| Benchmark | Gemini 1.5 Pro (May 2024) | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1296 | 1355 |
| ARC-AGI-2 | 0.8% | — |
| SimpleBench | 27.1% | — |
| Kagi LLM Benchmark | — | 62.3% |
| DTBench | 59% | — |
| BIG-Bench Hard | 89.2% | — |
| Epoch Capabilities Index | 131.73 | — |
| ForecastBench | 58.4 | — |
Math Step 3 leads
Gemini 1.5 Pro (May 2024): 25.8 (#266), Step 3: 37.6 (#148)
| Benchmark | Gemini 1.5 Pro (May 2024) | Step 3 |
|---|---|---|
| LMArena Math | 1315 | 1366 |
| OTIS Mock AIME 2024-2025 | 23.1% | — |
| Omni-MATH | 36.4% | — |
| MATH Level 5 | 70.4% | — |
Knowledge Step 3 leads
Gemini 1.5 Pro (May 2024): 29.4 (#239), Step 3: 36.8 (#164)
| Benchmark | Gemini 1.5 Pro (May 2024) | Step 3 |
|---|---|---|
| LMArena Expert | 1279 | 1333 |
| GPQA Diamond | 57.2% | — |
| Humanity's Last Exam | 4.6% | — |
| MMLU-Pro | 73.7% | — |
| Confabulations | 13.5% | — |
| GPQA (HELM) | 53.4% | — |
| MMLU | 86.9% | — |
Multimodal Gemini 1.5 Pro (May 2024) leads
Gemini 1.5 Pro (May 2024): 36.8 (#77), Step 3: 35.5 (#86)
| Benchmark | Gemini 1.5 Pro (May 2024) | Step 3 |
|---|---|---|
| LMArena Vision | 1161 | 1177 |
| Video-MME | 75% | — |
Multilingual Step 3 leads
Gemini 1.5 Pro (May 2024): 45.3 (#174), Step 3: 46.3 (#159)
| Benchmark | Gemini 1.5 Pro (May 2024) | Step 3 |
|---|---|---|
| LMArena Non-English | 1312 | 1327 |
| LMArena Chinese | 1331 | 1397 |
| LMArena German | 1286 | 1371 |
| LMArena Korean | 1298 | 1269 |
| LMArena Russian | 1320 | 1331 |
| LMArena Spanish | 1311 | 1371 |
| LMArena French | 1302 | — |
| LMArena Japanese | 1292 | — |
Instruction Following Step 3 leads
Gemini 1.5 Pro (May 2024): 68.6 (#185), Step 3: 70.4 (#164)
| Benchmark | Gemini 1.5 Pro (May 2024) | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1297 | 1332 |
| IFEval | 83.7% | — |
Long Context Too close to call
Gemini 1.5 Pro (May 2024): 39.8 (#169), Step 3: 40.3 (#157)
| Benchmark | Gemini 1.5 Pro (May 2024) | Step 3 |
|---|---|---|
| LMArena Longer Query | 1308 | 1326 |
Writing & Preference Step 3 leads
Gemini 1.5 Pro (May 2024): 52.4 (#172), Step 3: 54.3 (#151)
| Benchmark | Gemini 1.5 Pro (May 2024) | Step 3 |
|---|---|---|
| LMArena Text | 1319 | 1350 |
| LMArena Creative Writing | 1333 | 1321 |
| LMArena Multi-Turn | 1296 | 1341 |
| WildBench | 81.3% | — |
Frequently asked questions
Is Gemini 1.5 Pro (May 2024) better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 32.1 on the Noometry Index.
Is Gemini 1.5 Pro (May 2024) or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 34.2 in the Noometry coding category.
How many benchmarks do Gemini 1.5 Pro (May 2024) and Step 3 share?
16 benchmarks have published results for both models. Gemini 1.5 Pro (May 2024) has 45 scored results on Noometry and Step 3 has 17.