Model comparison
Gemma 4 31B IT vs Phi-4
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 31.2 on the Noometry Index. Phi-4 costs 1.7× less per token, which makes it the better buy when Gemma 4 31B IT's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Google
MicrosoftSummary
- They share 19 benchmarks with published results for both. Gemma 4 31B IT scores higher in 8 categories and Phi-4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 43.2 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.3% for Gemma 4 31B IT and 13.8% for Phi-4.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.09 / $0.34 for Gemma 4 31B IT.
- Gemma 4 31B IT accepts more context: 262K tokens versus 128K.
Side by side
| Gemma 4 31B IT | Phi-4 | |
|---|---|---|
| Provider | Microsoft | |
| Noometry Index | 43.5 | 31.2 |
| Released | 2026-04-02 | 2024-12-11 |
| Weights | Open | Open |
| Context window | 262K | 128K |
| Max output | 33K | 4K |
| Input $ / M tokens | $0.09 | $0.07 |
| Output $ / M tokens | $0.34 | $0.14 |
| Results tracked | 35 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Gemma 4 31B IT: 42.3 (#108), Phi-4: 34.4 (#239)
| Benchmark | Gemma 4 31B IT | Phi-4 |
|---|---|---|
| LMArena Coding | 1459 | 1231 |
| LMArena WebDev | 1366 | — |
| SciCode | 43.4% | — |
| WeirdML | 52.3% | — |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |
| ALE-Bench | 925.5 | — |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, Phi-4: 22.8 (#128)
| Benchmark | Gemma 4 31B IT | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), Phi-4: 17.7 (#291)
| Benchmark | Gemma 4 31B IT | Phi-4 |
|---|---|---|
| Chess Puzzles | 5% | 1% |
| LMArena Hard Prompts | 1448 | 1220 |
| Epoch Capabilities Index | 142.74 | 130.42 |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| CritPt | 1.4% | — |
| Thematic Generalization | 53% | — |
| LiveBench Reasoning | — | 47.8% |
| DTBench | 82.7% | — |
| LiveBench Data Analysis | — | 45.2% |
| LMCA | 39.3% | — |
| Surface Evolver Bench | 30.6% | — |
| LiveBench | — | 41.6% |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), Phi-4: 20.8 (#285)
| Benchmark | Gemma 4 31B IT | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 13.8% |
| LMArena Math | 1465 | 1246 |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
Knowledge Gemma 4 31B IT leads
Gemma 4 31B IT: 37.9 (#151), Phi-4: 32.6 (#209)
| Benchmark | Gemma 4 31B IT | Phi-4 |
|---|---|---|
| GPQA Diamond | 75.8% | 56.1% |
| Vectara Hallucination Rate | 7.4% | 3.7% |
| LMArena Expert | 1465 | 1203 |
| SimpleQA Verified | 10.4% | — |
| Confabulations | — | 29.4% |
| MMLU | — | 84.8% |
Multimodal Not comparable
Gemma 4 31B IT: 41.6 (#34), Phi-4: —
| Benchmark | Gemma 4 31B IT | Phi-4 |
|---|---|---|
| LMArena Vision | 1277 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), Phi-4: 37.2 (#237)
| Benchmark | Gemma 4 31B IT | Phi-4 |
|---|---|---|
| LMArena Non-English | 1431 | 1197 |
| LMArena Chinese | 1476 | 1212 |
| LMArena French | 1435 | 1224 |
| LMArena Russian | 1460 | 1209 |
| LMArena Spanish | 1444 | 1234 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1158 |
| LMArena Korean | — | 1151 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.5 (#61), Phi-4: 60.4 (#251)
| Benchmark | Gemma 4 31B IT | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#71), Phi-4: 36.9 (#226)
| Benchmark | Gemma 4 31B IT | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1446 | 1217 |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), Phi-4: 40.5 (#244)
| Benchmark | Gemma 4 31B IT | Phi-4 |
|---|---|---|
| LMArena Text | 1443 | 1217 |
| LMArena Creative Writing | 1415 | 1182 |
| LMArena Multi-Turn | 1452 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is Gemma 4 31B IT better than Phi-4?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 31.2 on the Noometry Index. Phi-4 costs 1.7× less per token, which makes it the better buy when Gemma 4 31B IT's lead doesn't matter for your workload.
Which is cheaper, Gemma 4 31B IT or Phi-4?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Gemma 4 31B IT lists at $0.09 and $0.34.
Is Gemma 4 31B IT or Phi-4 better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 128K.
How many benchmarks do Gemma 4 31B IT and Phi-4 share?
19 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Phi-4 has 37.