Model comparison
GPT-5.4 mini vs Phi-4
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 31.2 on the Noometry Index. Phi-4 costs 19× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
MicrosoftSummary
- They share 22 benchmarks with published results for both. GPT-5.4 mini scores higher in 9 categories and Phi-4 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 mini leads 45.5 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for GPT-5.4 mini and 13.8% for Phi-4.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
- GPT-5.4 mini accepts more context: 400K tokens versus 128K.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 mini | Phi-4 | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 45.0 | 31.2 |
| Released | 2026-03-17 | 2024-12-11 |
| Weights | Proprietary | Open |
| Context window | 400K | 128K |
| Max output | 128K | 4K |
| Input $ / M tokens | $0.75 | $0.07 |
| Output $ / M tokens | $4.50 | $0.14 |
| Results tracked | 46 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
GPT-5.4 mini: 45.2 (#72), Phi-4: 34.4 (#239)
| Benchmark | GPT-5.4 mini | Phi-4 |
|---|---|---|
| LMArena Coding | 1438 | 1231 |
| FrontierCode | 27% | — |
| LMArena WebDev | 1397 | — |
| SciCode | 49.9% | — |
| WeirdML | 60.3% | — |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |
| ALE-Bench | 1,189 | — |
Agentic & Tool Use GPT-5.4 mini leads
GPT-5.4 mini: 29.9 (#81), Phi-4: 22.8 (#128)
| Benchmark | GPT-5.4 mini | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| DeepResearch Bench | 36.3% | — |
| BALROG | — | 11.6% |
Reasoning GPT-5.4 mini leads
GPT-5.4 mini: 30.4 (#85), Phi-4: 17.7 (#291)
| Benchmark | GPT-5.4 mini | Phi-4 |
|---|---|---|
| Chess Puzzles | 24% | 1% |
| LMArena Hard Prompts | 1424 | 1220 |
| Epoch Capabilities Index | 148.84 | 130.42 |
| ARC-AGI-2 | 18.9% | — |
| Kagi LLM Benchmark | 37.9% | — |
| NYT Connections (extended) | 61.8% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | 10% | — |
| Thematic Generalization | 61.7% | — |
| LiveBench Reasoning | — | 47.8% |
| Mystery Game Puzzles | 11% | — |
| DTBench | 80% | — |
| LiveBench Data Analysis | — | 45.2% |
| LMCA | 40.8% | — |
| ForecastBench | 57 | — |
| LiveBench | — | 41.6% |
Math GPT-5.4 mini leads
GPT-5.4 mini: 45.5 (#75), Phi-4: 20.8 (#285)
| Benchmark | GPT-5.4 mini | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 13.8% |
| LMArena Math | 1419 | 1246 |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 9.8% | — |
| ProofBench | 21% | — |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
| FrontierMath (Feb 2025 set) | 28.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-5.4 mini leads
GPT-5.4 mini: 51.5 (#67), Phi-4: 32.6 (#209)
| Benchmark | GPT-5.4 mini | Phi-4 |
|---|---|---|
| GPQA Diamond | 86.9% | 56.1% |
| Vectara Hallucination Rate | 5.5% | 3.7% |
| LMArena Expert | 1435 | 1203 |
| SimpleQA Verified | 29.4% | — |
| Confabulations | — | 29.4% |
| MMLU | — | 84.8% |
Multimodal Not comparable
GPT-5.4 mini: 39.7 (#56), Phi-4: —
| Benchmark | GPT-5.4 mini | Phi-4 |
|---|---|---|
| LMArena Vision | 1245 | — |
Multilingual GPT-5.4 mini leads
GPT-5.4 mini: 51.9 (#96), Phi-4: 37.2 (#237)
| Benchmark | GPT-5.4 mini | Phi-4 |
|---|---|---|
| LMArena Non-English | 1405 | 1197 |
| LMArena Chinese | 1446 | 1212 |
| LMArena French | 1440 | 1224 |
| LMArena German | 1409 | 1222 |
| LMArena Japanese | 1374 | 1158 |
| LMArena Korean | 1368 | 1151 |
| LMArena Russian | 1417 | 1209 |
| LMArena Spanish | 1405 | 1234 |
Instruction Following GPT-5.4 mini leads
GPT-5.4 mini: 74.1 (#102), Phi-4: 60.4 (#251)
| Benchmark | GPT-5.4 mini | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1405 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
Long Context GPT-5.4 mini leads
GPT-5.4 mini: 43.0 (#112), Phi-4: 36.9 (#226)
| Benchmark | GPT-5.4 mini | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1407 | 1217 |
Writing & Preference GPT-5.4 mini leads
GPT-5.4 mini: 64.0 (#58), Phi-4: 40.5 (#244)
| Benchmark | GPT-5.4 mini | Phi-4 |
|---|---|---|
| LMArena Text | 1412 | 1217 |
| LMArena Creative Writing | 1370 | 1182 |
| LMArena Multi-Turn | 1429 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| EQ-Bench Creative Writing | 1665 | — |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is GPT-5.4 mini better than Phi-4?
GPT-5.4 mini is the stronger model overall, scoring 45.0 to 31.2 on the Noometry Index. Phi-4 costs 19× less per token, which makes it the better buy when GPT-5.4 mini's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 mini or Phi-4?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; GPT-5.4 mini lists at $0.75 and $4.50.
Is GPT-5.4 mini or Phi-4 better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 mini does, with 400K tokens against 128K.
How many benchmarks do GPT-5.4 mini and Phi-4 share?
22 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Phi-4 has 37.