Model comparison
Phi-4 vs Step 3.5 Flash
Step 3.5 Flash is the stronger model overall, scoring 41.7 to 30.9 on the Noometry Index. Phi-4 costs 1.7× less per token, which makes it the better buy when Step 3.5 Flash's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Microsoft
StepFunSummary
- They share 17 benchmarks with published results for both. Phi-4 scores higher in 0 categories and Step 3.5 Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 3.5 Flash leads 40.6 to 19.6.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.10 / $0.30 for Step 3.5 Flash.
- Step 3.5 Flash accepts more context: 256K tokens versus 128K.
Side by side
| Phi-4 | Step 3.5 Flash | |
|---|---|---|
| Provider | Microsoft | StepFun |
| Noometry Index | 30.9 | 41.7 |
| Released | 2024-12-11 | 2026-01-29 |
| Weights | Open | Open |
| Context window | 128K | 256K |
| Max output | 4K | 256K |
| Input $ / M tokens | $0.07 | $0.10 |
| Output $ / M tokens | $0.14 | $0.30 |
| Results tracked | 37 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3.5 Flash leads
Phi-4: 33.8 (#237), Step 3.5 Flash: 41.7 (#106)
| Benchmark | Phi-4 | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1231 | 1436 |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 30.7% | — |
| BigCodeBench Complete | 55.4% | — |
Agentic & Tool Use Not comparable
Phi-4: 22.6 (#120), Step 3.5 Flash: —
| Benchmark | Phi-4 | Step 3.5 Flash |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.8% | — |
| BALROG | 11.6% | — |
Reasoning Step 3.5 Flash leads
Phi-4: 17.0 (#287), Step 3.5 Flash: 21.2 (#202)
| Benchmark | Phi-4 | Step 3.5 Flash |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1411 |
| NYT Connections (extended) | — | 28.4% |
| Chess Puzzles | 1% | — |
| LiveBench Reasoning | 47.8% | — |
| LiveBench Data Analysis | 45.2% | — |
| Epoch Capabilities Index | 130.43 | — |
| LiveBench | 41.6% | — |
Math Step 3.5 Flash leads
Phi-4: 19.6 (#279), Step 3.5 Flash: 40.6 (#77)
| Benchmark | Phi-4 | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1246 | 1408 |
| MathArena Final-Answer Competitions | — | 66.8% |
| OTIS Mock AIME 2024-2025 | 13.8% | — |
| LiveBench Math | 42% | — |
| MATH Level 5 | 64.9% | — |
Knowledge Step 3.5 Flash leads
Phi-4: 32.5 (#204), Step 3.5 Flash: 39.0 (#127)
| Benchmark | Phi-4 | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1203 | 1422 |
| GPQA Diamond | 56.1% | — |
| Confabulations | 29.4% | — |
| Vectara Hallucination Rate | 3.7% | — |
| MMLU | 84.8% | — |
Multilingual Step 3.5 Flash leads
Phi-4: 37.4 (#232), Step 3.5 Flash: 50.7 (#115)
| Benchmark | Phi-4 | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1197 | 1386 |
| LMArena Chinese | 1211 | 1448 |
| LMArena French | 1223 | 1421 |
| LMArena German | 1222 | 1404 |
| LMArena Japanese | 1158 | 1353 |
| LMArena Korean | 1151 | 1350 |
| LMArena Russian | 1209 | 1385 |
| LMArena Spanish | 1234 | 1419 |
Instruction Following Step 3.5 Flash leads
Phi-4: 60.6 (#246), Step 3.5 Flash: 73.3 (#119)
| Benchmark | Phi-4 | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1201 | 1385 |
| LiveBench Instruction Following | 58.4% | — |
Long Context Step 3.5 Flash leads
Phi-4: 36.9 (#221), Step 3.5 Flash: 42.8 (#112)
| Benchmark | Phi-4 | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1217 | 1402 |
Writing & Preference Step 3.5 Flash leads
Phi-4: 40.7 (#239), Step 3.5 Flash: 59.0 (#108)
| Benchmark | Phi-4 | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1217 | 1403 |
| LMArena Creative Writing | 1182 | 1357 |
| LMArena Multi-Turn | 1206 | 1405 |
| Short-Story Creative Writing | 62.6% | — |
| LiveBench Language | 25.6% | — |
Frequently asked questions
Is Phi-4 better than Step 3.5 Flash?
Step 3.5 Flash is the stronger model overall, scoring 41.7 to 30.9 on the Noometry Index. Phi-4 costs 1.7× less per token, which makes it the better buy when Step 3.5 Flash's lead doesn't matter for your workload.
Which is cheaper, Phi-4 or Step 3.5 Flash?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Step 3.5 Flash lists at $0.10 and $0.30.
Is Phi-4 or Step 3.5 Flash better for coding?
Step 3.5 Flash scores higher on coding benchmarks: 41.7 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
Step 3.5 Flash does, with 256K tokens against 128K.
How many benchmarks do Phi-4 and Step 3.5 Flash share?
17 benchmarks have published results for both models. Phi-4 has 37 scored results on Noometry and Step 3.5 Flash has 19.