Model comparison
Phi 3 Mini 4k Instruct vs Step 3
Step 3 is the stronger model overall, scoring 39.9 to 27.4 on the Noometry Index.
Last verified . 15 shared benchmarks.
Microsoft
StepFunSummary
- They share 15 benchmarks with published results for both. Phi 3 Mini 4k Instruct scores higher in 0 categories and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Step 3 leads 54.5 to 27.7.
Side by side
| Phi 3 Mini 4k Instruct | Step 3 | |
|---|---|---|
| Provider | Microsoft | StepFun |
| Noometry Index | 27.4 | 39.9 |
| Released | 2024-04-23 | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Phi 3 Mini 4k Instruct: 26.1 (#319), Step 3: 39.5 (#144)
| Benchmark | Phi 3 Mini 4k Instruct | Step 3 |
|---|---|---|
| LMArena Coding | 1093 | 1367 |
| LiveBench Coding | 15.5% | — |
| HumanEval+ | 59.1% | — |
| MBPP+ | 54.2% | — |
Reasoning Step 3 leads
Phi 3 Mini 4k Instruct: 13.6 (#326), Step 3: 27.6 (#102)
| Benchmark | Phi 3 Mini 4k Instruct | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1072 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 26.8% | — |
| LiveBench Data Analysis | 34.7% | — |
| Adversarial NLI | 52.8% | — |
| BIG-Bench Hard | 71.7% | — |
| HellaSwag | 76.7% | — |
| LiveBench | 22.4% | — |
| WinoGrande | 70.8% | — |
Math Step 3 leads
Phi 3 Mini 4k Instruct: 24.5 (#258), Step 3: 34.9 (#148)
| Benchmark | Phi 3 Mini 4k Instruct | Step 3 |
|---|---|---|
| LMArena Math | 1111 | 1366 |
| LiveBench Math | 15.7% | — |
Knowledge Step 3 leads
Phi 3 Mini 4k Instruct: 27.9 (#241), Step 3: 36.2 (#161)
| Benchmark | Phi 3 Mini 4k Instruct | Step 3 |
|---|---|---|
| LMArena Expert | 1045 | 1333 |
| ARC (AI2) Challenge | 84.9% | — |
| MMLU | 68.8% | — |
| OpenBookQA | 88% | — |
| TriviaQA | 64% | — |
Multimodal Not comparable
Phi 3 Mini 4k Instruct: —, Step 3: 34.9 (#81)
| Benchmark | Phi 3 Mini 4k Instruct | Step 3 |
|---|---|---|
| LMArena Vision | — | 1176 |
Multilingual Step 3 leads
Phi 3 Mini 4k Instruct: 26.5 (#275), Step 3: 46.5 (#154)
| Benchmark | Phi 3 Mini 4k Instruct | Step 3 |
|---|---|---|
| LMArena Non-English | 1021 | 1327 |
| LMArena Chinese | 1021 | 1396 |
| LMArena German | 1044 | 1371 |
| LMArena Korean | 905 | 1271 |
| LMArena Russian | 1022 | 1331 |
| LMArena Spanish | 1084 | 1372 |
| LMArena French | 1075 | — |
| LMArena Japanese | 935 | — |
Instruction Following Step 3 leads
Phi 3 Mini 4k Instruct: 48.0 (#298), Step 3: 70.6 (#159)
| Benchmark | Phi 3 Mini 4k Instruct | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1053 | 1332 |
| LiveBench Instruction Following | 39.1% | — |
Long Context Step 3 leads
Phi 3 Mini 4k Instruct: 31.7 (#271), Step 3: 40.3 (#152)
| Benchmark | Phi 3 Mini 4k Instruct | Step 3 |
|---|---|---|
| LMArena Longer Query | 1044 | 1326 |
Writing & Preference Step 3 leads
Phi 3 Mini 4k Instruct: 27.7 (#295), Step 3: 54.5 (#146)
| Benchmark | Phi 3 Mini 4k Instruct | Step 3 |
|---|---|---|
| LMArena Text | 1073 | 1351 |
| LMArena Creative Writing | 1037 | 1322 |
| LMArena Multi-Turn | 1018 | 1340 |
| LiveBench Language | 9.2% | — |
Frequently asked questions
Is Phi 3 Mini 4k Instruct better than Step 3?
Step 3 is the stronger model overall, scoring 39.9 to 27.4 on the Noometry Index.
Is Phi 3 Mini 4k Instruct or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 39.5 versus 26.1 in the Noometry coding category.
How many benchmarks do Phi 3 Mini 4k Instruct and Step 3 share?
15 benchmarks have published results for both models. Phi 3 Mini 4k Instruct has 35 scored results on Noometry and Step 3 has 17.