Model comparison
Phi-4 vs Solar Pro4
Solar Pro4 is the stronger model overall, scoring 41.5 to 30.9 on the Noometry Index. Phi-4 costs 6.0× less per token, which makes it the better buy when Solar Pro4's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Microsoft
UpstageSummary
- They share 17 benchmarks with published results for both. Phi-4 scores higher in 0 categories and Solar Pro4 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Solar Pro4 leads 36.1 to 19.6.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.30 / $1.20 for Solar Pro4.
- Solar Pro4 accepts more context: 524K tokens versus 128K.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Phi-4 | Solar Pro4 | |
|---|---|---|
| Provider | Microsoft | Upstage |
| Noometry Index | 30.9 | 41.5 |
| Released | 2024-12-11 | 2026-08-06 |
| Weights | Open | Proprietary |
| Context window | 128K | 524K |
| Max output | 4K | 131K |
| Input $ / M tokens | $0.07 | $0.30 |
| Output $ / M tokens | $0.14 | $1.20 |
| Results tracked | 37 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Solar Pro4 leads
Phi-4: 33.8 (#237), Solar Pro4: 39.9 (#135)
| Benchmark | Phi-4 | Solar Pro4 |
|---|---|---|
| LMArena Coding | 1231 | 1439 |
| LMArena WebDev | — | 1371 |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 30.7% | — |
| BigCodeBench Complete | 55.4% | — |
Agentic & Tool Use Not comparable
Phi-4: 22.6 (#120), Solar Pro4: —
| Benchmark | Phi-4 | Solar Pro4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.8% | — |
| BALROG | 11.6% | — |
Reasoning Solar Pro4 leads
Phi-4: 17.0 (#287), Solar Pro4: 27.3 (#107)
| Benchmark | Phi-4 | Solar Pro4 |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1398 |
| Chess Puzzles | 1% | — |
| LiveBench Reasoning | 47.8% | — |
| LiveBench Data Analysis | 45.2% | — |
| Epoch Capabilities Index | 130.43 | — |
| LiveBench | 41.6% | — |
Math Solar Pro4 leads
Phi-4: 19.6 (#279), Solar Pro4: 36.1 (#129)
| Benchmark | Phi-4 | Solar Pro4 |
|---|---|---|
| LMArena Math | 1246 | 1418 |
| OTIS Mock AIME 2024-2025 | 13.8% | — |
| LiveBench Math | 42% | — |
| MATH Level 5 | 64.9% | — |
Knowledge Solar Pro4 leads
Phi-4: 32.5 (#204), Solar Pro4: 39.0 (#128)
| Benchmark | Phi-4 | Solar Pro4 |
|---|---|---|
| LMArena Expert | 1203 | 1421 |
| GPQA Diamond | 56.1% | — |
| Confabulations | 29.4% | — |
| Vectara Hallucination Rate | 3.7% | — |
| MMLU | 84.8% | — |
Multilingual Solar Pro4 leads
Phi-4: 37.4 (#232), Solar Pro4: 49.0 (#134)
| Benchmark | Phi-4 | Solar Pro4 |
|---|---|---|
| LMArena Non-English | 1197 | 1362 |
| LMArena Chinese | 1211 | 1417 |
| LMArena French | 1223 | 1403 |
| LMArena German | 1222 | 1351 |
| LMArena Japanese | 1158 | 1300 |
| LMArena Korean | 1151 | 1383 |
| LMArena Russian | 1209 | 1360 |
| LMArena Spanish | 1234 | 1408 |
Instruction Following Solar Pro4 leads
Phi-4: 60.6 (#246), Solar Pro4: 72.8 (#128)
| Benchmark | Phi-4 | Solar Pro4 |
|---|---|---|
| LMArena Instruction Following | 1201 | 1376 |
| LiveBench Instruction Following | 58.4% | — |
Long Context Solar Pro4 leads
Phi-4: 36.9 (#221), Solar Pro4: 42.0 (#125)
| Benchmark | Phi-4 | Solar Pro4 |
|---|---|---|
| LMArena Longer Query | 1217 | 1379 |
Writing & Preference Solar Pro4 leads
Phi-4: 40.7 (#239), Solar Pro4: 56.7 (#133)
| Benchmark | Phi-4 | Solar Pro4 |
|---|---|---|
| LMArena Text | 1217 | 1386 |
| LMArena Creative Writing | 1182 | 1313 |
| LMArena Multi-Turn | 1206 | 1391 |
| Short-Story Creative Writing | 62.6% | — |
| LiveBench Language | 25.6% | — |
Frequently asked questions
Is Phi-4 better than Solar Pro4?
Solar Pro4 is the stronger model overall, scoring 41.5 to 30.9 on the Noometry Index. Phi-4 costs 6.0× less per token, which makes it the better buy when Solar Pro4's lead doesn't matter for your workload.
Which is cheaper, Phi-4 or Solar Pro4?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Solar Pro4 lists at $0.30 and $1.20.
Is Phi-4 or Solar Pro4 better for coding?
Solar Pro4 scores higher on coding benchmarks: 39.9 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
Solar Pro4 does, with 524K tokens against 128K.
How many benchmarks do Phi-4 and Solar Pro4 share?
17 benchmarks have published results for both models. Phi-4 has 37 scored results on Noometry and Solar Pro4 has 18.