Model comparison
Claude Sonnet 5 vs Phi-4
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 31.2 on the Noometry Index. Phi-4 costs 46× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
MicrosoftSummary
- They share 21 benchmarks with published results for both. Claude Sonnet 5 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 Claude Sonnet 5 leads 66.2 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 94.7% for Claude Sonnet 5 and 13.8% for Phi-4.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
- Claude Sonnet 5 accepts more context: 1M tokens versus 128K.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 5 | Phi-4 | |
|---|---|---|
| Provider | Anthropic | Microsoft |
| Noometry Index | 54.6 | 31.2 |
| Released | 2026-06-29 | 2024-12-11 |
| Weights | Proprietary | Open |
| Context window | 1M | 128K |
| Max output | 128K | 4K |
| Input $ / M tokens | $2 | $0.07 |
| Output $ / M tokens | $10 | $0.14 |
| Results tracked | 51 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 5 leads
Claude Sonnet 5: 55.5 (#26), Phi-4: 34.4 (#239)
| Benchmark | Claude Sonnet 5 | Phi-4 |
|---|---|---|
| LMArena Coding | 1483 | 1231 |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.7% | — |
| CursorBench | 34.1% | — |
| LMArena WebDev | 1541 | — |
| SciCode | 54.3% | — |
| GSO | 37.3% | — |
| WeirdML | 68.8% | — |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |
| ALE-Bench | 1,463 | — |
Agentic & Tool Use Claude Sonnet 5 leads
Claude Sonnet 5: 42.8 (#18), Phi-4: 22.8 (#128)
| Benchmark | Claude Sonnet 5 | Phi-4 |
|---|---|---|
| APEX-Agents | 54.5% | — |
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
| GBAEval | 65.3% | — |
| LMArena Search | 1194 | — |
| Vending-Bench 2 | 6,378 | — |
Reasoning Claude Sonnet 5 leads
Claude Sonnet 5: 49.1 (#39), Phi-4: 17.7 (#291)
| Benchmark | Claude Sonnet 5 | Phi-4 |
|---|---|---|
| Chess Puzzles | 35% | 1% |
| LMArena Hard Prompts | 1461 | 1220 |
| Epoch Capabilities Index | 156.21 | 130.42 |
| SimpleBench | 60.6% | — |
| NYT Connections (extended) | 75.1% | — |
| CritPt | 16.9% | — |
| LiveBench Reasoning | — | 47.8% |
| Mystery Game Puzzles | 35% | — |
| DTBench | 92.5% | — |
| LiveBench Data Analysis | — | 45.2% |
| LMCA | 50% | — |
| Surface Evolver Bench | 60% | — |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | 61.1 | — |
| LiveBench | — | 41.6% |
Math Claude Sonnet 5 leads
Claude Sonnet 5: 66.2 (#27), Phi-4: 20.8 (#285)
| Benchmark | Claude Sonnet 5 | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 94.7% | 13.8% |
| LMArena Math | 1467 | 1246 |
| FrontierMath (Tiers 1-3) | 65.6% | — |
| FrontierMath Tier 4 | 29.3% | — |
| ProofBench | 77% | — |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
Knowledge Claude Sonnet 5 leads
Claude Sonnet 5: 55.6 (#47), Phi-4: 32.6 (#209)
| Benchmark | Claude Sonnet 5 | Phi-4 |
|---|---|---|
| GPQA Diamond | 90.5% | 56.1% |
| LMArena Expert | 1490 | 1203 |
| SimpleQA Verified | 33.7% | — |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| MMLU | — | 84.8% |
Multimodal Not comparable
Claude Sonnet 5: 42.4 (#31), Phi-4: —
| Benchmark | Claude Sonnet 5 | Phi-4 |
|---|---|---|
| LMArena Vision | 1274 | — |
| Blueprint-Bench 2 | 24.9% | — |
| LMArena Document | 1466 | — |
Multilingual Claude Sonnet 5 leads
Claude Sonnet 5: 53.8 (#55), Phi-4: 37.2 (#237)
| Benchmark | Claude Sonnet 5 | Phi-4 |
|---|---|---|
| LMArena Non-English | 1431 | 1197 |
| LMArena Chinese | 1477 | 1212 |
| LMArena French | 1460 | 1224 |
| LMArena German | 1440 | 1222 |
| LMArena Japanese | 1422 | 1158 |
| LMArena Korean | 1411 | 1151 |
| LMArena Russian | 1451 | 1209 |
| LMArena Spanish | 1437 | 1234 |
Instruction Following Claude Sonnet 5 leads
Claude Sonnet 5: 76.3 (#41), Phi-4: 60.4 (#251)
| Benchmark | Claude Sonnet 5 | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1452 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
Long Context Claude Sonnet 5 leads
Claude Sonnet 5: 44.8 (#55), Phi-4: 36.9 (#226)
| Benchmark | Claude Sonnet 5 | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1463 | 1217 |
Writing & Preference Claude Sonnet 5 leads
Claude Sonnet 5: 69.2 (#25), Phi-4: 40.5 (#244)
| Benchmark | Claude Sonnet 5 | Phi-4 |
|---|---|---|
| LMArena Text | 1442 | 1217 |
| LMArena Creative Writing | 1416 | 1182 |
| LMArena Multi-Turn | 1454 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| EQ-Bench Creative Writing | 1794 | — |
| EQ-Bench 4 | 1236 | — |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is Claude Sonnet 5 better than Phi-4?
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 31.2 on the Noometry Index. Phi-4 costs 46× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.
Which is cheaper, Claude Sonnet 5 or Phi-4?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Claude Sonnet 5 lists at $2 and $10.
Is Claude Sonnet 5 or Phi-4 better for coding?
Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 5 does, with 1M tokens against 128K.
How many benchmarks do Claude Sonnet 5 and Phi-4 share?
21 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and Phi-4 has 37.