Model comparison
Llama 4 Maverick vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 30.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
Meta
StepFunSummary
- They share 17 benchmarks with published results for both. Llama 4 Maverick scores higher in 1 category and Step 3 in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 3 leads 28.4 to 10.1.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 55.9% for Llama 4 Maverick and 62.3% for Step 3.
Side by side
| Llama 4 Maverick | Step 3 | |
|---|---|---|
| Provider | Meta | StepFun |
| Noometry Index | 30.9 | 40.5 |
| Released | 2025-04-05 | — |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.19 | — |
| Output $ / M tokens | $0.65 | — |
| Results tracked | 54 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Llama 4 Maverick: 26.6 (#324), Step 3: 40.1 (#147)
| Benchmark | Llama 4 Maverick | Step 3 |
|---|---|---|
| LMArena Coding | 1302 | 1367 |
| SWE-bench Verified (bash only) | 21% | — |
| Aider Polyglot | 15.6% | — |
| SciCode | 33.1% | — |
| WeirdML | 24.5% | — |
| BigCodeBench Instruct | 49.7% | — |
| BigCodeBench Complete | 61.4% | — |
| ALE-Bench | 172.97 | — |
Agentic & Tool Use Not comparable
Llama 4 Maverick: 28.2 (#91), Step 3: —
| Benchmark | Llama 4 Maverick | Step 3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | — |
Reasoning Step 3 leads
Llama 4 Maverick: 10.1 (#342), Step 3: 28.4 (#105)
| Benchmark | Llama 4 Maverick | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 55.9% | 62.3% |
| LMArena Hard Prompts | 1281 | 1355 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 27.7% | — |
| NYT Connections (extended) | 8% | — |
| ARC-AGI-1 | 4.4% | — |
| CritPt | 0% | — |
| EnigmaEval | 0.6% | — |
| DTBench | 61.9% | — |
| LMCA | 15.9% | — |
| Epoch Capabilities Index | 132.2 | — |
| ForecastBench | 57.5 | — |
Math Step 3 leads
Llama 4 Maverick: 26.0 (#262), Step 3: 37.6 (#148)
| Benchmark | Llama 4 Maverick | Step 3 |
|---|---|---|
| LMArena Math | 1299 | 1366 |
| OTIS Mock AIME 2024-2025 | 20.6% | — |
| Omni-MATH | 42.2% | — |
| MATH Level 5 | 73% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |
Knowledge Step 3 leads
Llama 4 Maverick: 33.4 (#204), Step 3: 36.8 (#164)
| Benchmark | Llama 4 Maverick | Step 3 |
|---|---|---|
| LMArena Expert | 1259 | 1333 |
| GPQA Diamond | 67% | — |
| Humanity's Last Exam | 5.7% | — |
| MMLU-Pro | 81% | — |
| Confabulations | 22.6% | — |
| Vectara Hallucination Rate | 8.2% | — |
| GPQA (HELM) | 65% | — |
Multimodal Step 3 leads
Llama 4 Maverick: 31.6 (#105), Step 3: 35.5 (#86)
| Benchmark | Llama 4 Maverick | Step 3 |
|---|---|---|
| LMArena Vision | 1142 | 1177 |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual Step 3 leads
Llama 4 Maverick: 42.2 (#195), Step 3: 46.3 (#159)
| Benchmark | Llama 4 Maverick | Step 3 |
|---|---|---|
| LMArena Non-English | 1269 | 1327 |
| LMArena Chinese | 1277 | 1397 |
| LMArena German | 1291 | 1371 |
| LMArena Korean | 1203 | 1269 |
| LMArena Russian | 1286 | 1331 |
| LMArena Spanish | 1293 | 1371 |
| LMArena French | 1259 | — |
| LMArena Japanese | 1207 | — |
Instruction Following Llama 4 Maverick leads
Llama 4 Maverick: 71.7 (#146), Step 3: 70.4 (#164)
| Benchmark | Llama 4 Maverick | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1267 | 1332 |
| IFEval | 90.8% | — |
Long Context Step 3 leads
Llama 4 Maverick: 31.4 (#279), Step 3: 40.3 (#157)
| Benchmark | Llama 4 Maverick | Step 3 |
|---|---|---|
| LMArena Longer Query | 1280 | 1326 |
| Fiction.LiveBench | 46.2% | — |
Writing & Preference Step 3 leads
Llama 4 Maverick: 38.8 (#252), Step 3: 54.3 (#151)
| Benchmark | Llama 4 Maverick | Step 3 |
|---|---|---|
| LMArena Text | 1287 | 1350 |
| LMArena Creative Writing | 1267 | 1321 |
| LMArena Multi-Turn | 1289 | 1341 |
| Short-Story Creative Writing | 62% | — |
| EQ-Bench Creative Writing | 860 | — |
| WildBench | 80% | — |
Frequently asked questions
Is Llama 4 Maverick better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 30.9 on the Noometry Index.
Is Llama 4 Maverick or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 26.6 in the Noometry coding category.
How many benchmarks do Llama 4 Maverick and Step 3 share?
17 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and Step 3 has 17.