Model comparison
Phi-4 vs Qwen3.6 Plus
Qwen3.6 Plus is the stronger model overall, scoring 47.5 to 31.2 on the Noometry Index. Phi-4 costs 13× less per token, which makes it the better buy when Qwen3.6 Plus's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 21 benchmarks with published results for both. Phi-4 scores higher in 0 categories and Qwen3.6 Plus in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.6 Plus leads 51.8 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 13.8% for Phi-4 and 93.3% for Qwen3.6 Plus.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.50 / $3 for Qwen3.6 Plus.
- Qwen3.6 Plus accepts more context: 1M tokens versus 128K.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Phi-4 | Qwen3.6 Plus | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 31.2 | 47.5 |
| Released | 2024-12-11 | 2026-03-31 |
| Weights | Open | Proprietary |
| Context window | 128K | 1M |
| Max output | 4K | 66K |
| Input $ / M tokens | $0.07 | $0.50 |
| Output $ / M tokens | $0.14 | $3 |
| Results tracked | 37 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.6 Plus leads
Phi-4: 34.4 (#239), Qwen3.6 Plus: 40.8 (#130)
| Benchmark | Phi-4 | Qwen3.6 Plus |
|---|---|---|
| LMArena Coding | 1231 | 1467 |
| SWE-bench Verified | — | 57.9% |
| LMArena WebDev | — | 1461 |
| SciCode | — | 40.7% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 30.7% | — |
| BigCodeBench Complete | 55.4% | — |
| ALE-Bench | — | 670.15 |
Agentic & Tool Use Not comparable
Phi-4: 22.8 (#128), Qwen3.6 Plus: —
| Benchmark | Phi-4 | Qwen3.6 Plus |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.8% | — |
| BALROG | 11.6% | — |
| Vending-Bench 2 | — | 5,115 |
Reasoning Qwen3.6 Plus leads
Phi-4: 17.7 (#291), Qwen3.6 Plus: 29.3 (#93)
| Benchmark | Phi-4 | Qwen3.6 Plus |
|---|---|---|
| Chess Puzzles | 1% | 17% |
| LMArena Hard Prompts | 1220 | 1449 |
| Epoch Capabilities Index | 130.42 | 147.65 |
| NYT Connections (extended) | — | 60.3% |
| CritPt | — | 2.9% |
| Thematic Generalization | — | 59.5% |
| LiveBench Reasoning | 47.8% | — |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 81.9% |
| LiveBench Data Analysis | 45.2% | — |
| LMCA | — | 33.1% |
| LiveBench | 41.6% | — |
Math Qwen3.6 Plus leads
Phi-4: 20.8 (#285), Qwen3.6 Plus: 51.8 (#54)
| Benchmark | Phi-4 | Qwen3.6 Plus |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 13.8% | 93.3% |
| LMArena Math | 1246 | 1450 |
| FrontierMath (Tiers 1-3) | — | 38.2% |
| LiveBench Math | 42% | — |
| MATH Level 5 | 64.9% | — |
| FrontierMath (Feb 2025 set) | — | 26.2% |
| FrontierMath Tier 4 (v1) | — | 8.3% |
Knowledge Qwen3.6 Plus leads
Phi-4: 32.6 (#209), Qwen3.6 Plus: 56.1 (#45)
| Benchmark | Phi-4 | Qwen3.6 Plus |
|---|---|---|
| GPQA Diamond | 56.1% | 88.4% |
| LMArena Expert | 1203 | 1454 |
| SimpleQA Verified | — | 44.1% |
| Confabulations | 29.4% | — |
| Vectara Hallucination Rate | 3.7% | — |
| MMLU | 84.8% | — |
Multilingual Qwen3.6 Plus leads
Phi-4: 37.2 (#237), Qwen3.6 Plus: 53.3 (#70)
| Benchmark | Phi-4 | Qwen3.6 Plus |
|---|---|---|
| LMArena Non-English | 1197 | 1424 |
| LMArena Chinese | 1212 | 1477 |
| LMArena French | 1224 | 1455 |
| LMArena German | 1222 | 1452 |
| LMArena Japanese | 1158 | 1389 |
| LMArena Korean | 1151 | 1379 |
| LMArena Russian | 1209 | 1434 |
| LMArena Spanish | 1234 | 1432 |
Instruction Following Qwen3.6 Plus leads
Phi-4: 60.4 (#251), Qwen3.6 Plus: 75.0 (#74)
| Benchmark | Phi-4 | Qwen3.6 Plus |
|---|---|---|
| LMArena Instruction Following | 1201 | 1425 |
| LiveBench Instruction Following | 58.4% | — |
Long Context Qwen3.6 Plus leads
Phi-4: 36.9 (#226), Qwen3.6 Plus: 45.2 (#49)
| Benchmark | Phi-4 | Qwen3.6 Plus |
|---|---|---|
| LMArena Longer Query | 1217 | 1439 |
| CL-bench | — | 20.3% |
Writing & Preference Qwen3.6 Plus leads
Phi-4: 40.5 (#244), Qwen3.6 Plus: 62.2 (#82)
| Benchmark | Phi-4 | Qwen3.6 Plus |
|---|---|---|
| LMArena Text | 1217 | 1437 |
| LMArena Creative Writing | 1182 | 1404 |
| LMArena Multi-Turn | 1206 | 1438 |
| Short-Story Creative Writing | 62.6% | — |
| LiveBench Language | 25.6% | — |
Frequently asked questions
Is Phi-4 better than Qwen3.6 Plus?
Qwen3.6 Plus is the stronger model overall, scoring 47.5 to 31.2 on the Noometry Index. Phi-4 costs 13× less per token, which makes it the better buy when Qwen3.6 Plus's lead doesn't matter for your workload.
Which is cheaper, Phi-4 or Qwen3.6 Plus?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Qwen3.6 Plus lists at $0.50 and $3.
Is Phi-4 or Qwen3.6 Plus better for coding?
Qwen3.6 Plus scores higher on coding benchmarks: 40.8 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
Qwen3.6 Plus does, with 1M tokens against 128K.
How many benchmarks do Phi-4 and Qwen3.6 Plus share?
21 benchmarks have published results for both models. Phi-4 has 37 scored results on Noometry and Qwen3.6 Plus has 37.