Model comparison
DeepSeek-R1-Distill-Qwen-1.5B vs Phi-4
Phi-4 is the stronger model overall, scoring 31.2 to 26.1 on the Noometry Index.
Last verified . 5 shared benchmarks.
DeepSeek
MicrosoftSummary
- They share 5 benchmarks with published results for both. DeepSeek-R1-Distill-Qwen-1.5B scores higher in 2 categories and Phi-4 in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Phi-4 leads 32.6 to 16.0.
- The biggest single-benchmark swing is BigCodeBench Complete: 7.9% for DeepSeek-R1-Distill-Qwen-1.5B and 55.4% for Phi-4.
Side by side
| DeepSeek-R1-Distill-Qwen-1.5B | Phi-4 | |
|---|---|---|
| Provider | DeepSeek | Microsoft |
| Noometry Index | 26.1 | 31.2 |
| Released | 2025-01-20 | 2024-12-11 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.07 |
| Output $ / M tokens | — | $0.14 |
| Results tracked | 5 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi-4 leads
DeepSeek-R1-Distill-Qwen-1.5B: 21.8 (#336), Phi-4: 34.4 (#239)
| Benchmark | DeepSeek-R1-Distill-Qwen-1.5B | Phi-4 |
|---|---|---|
| BigCodeBench Instruct | 7% | 45.5% |
| BigCodeBench Complete | 7.9% | 55.4% |
| LiveBench Coding | — | 30.7% |
| LMArena Coding | — | 1231 |
Agentic & Tool Use Not comparable
DeepSeek-R1-Distill-Qwen-1.5B: —, Phi-4: 22.8 (#128)
| Benchmark | DeepSeek-R1-Distill-Qwen-1.5B | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
Reasoning DeepSeek-R1-Distill-Qwen-1.5B leads
DeepSeek-R1-Distill-Qwen-1.5B: 19.2 (#262), Phi-4: 17.7 (#291)
| Benchmark | DeepSeek-R1-Distill-Qwen-1.5B | Phi-4 |
|---|---|---|
| Chess Puzzles | 0% | 1% |
| LiveBench Reasoning | — | 47.8% |
| LMArena Hard Prompts | — | 1220 |
| LiveBench Data Analysis | — | 45.2% |
| Epoch Capabilities Index | — | 130.42 |
| LiveBench | — | 41.6% |
Math DeepSeek-R1-Distill-Qwen-1.5B leads
DeepSeek-R1-Distill-Qwen-1.5B: 23.0 (#274), Phi-4: 20.8 (#285)
| Benchmark | DeepSeek-R1-Distill-Qwen-1.5B | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 21.4% | 13.8% |
| LiveBench Math | — | 42% |
| LMArena Math | — | 1246 |
| MATH Level 5 | — | 64.9% |
Knowledge Phi-4 leads
DeepSeek-R1-Distill-Qwen-1.5B: 16.0 (#290), Phi-4: 32.6 (#209)
| Benchmark | DeepSeek-R1-Distill-Qwen-1.5B | Phi-4 |
|---|---|---|
| GPQA Diamond | 33.6% | 56.1% |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| LMArena Expert | — | 1203 |
| MMLU | — | 84.8% |
Multilingual Not comparable
DeepSeek-R1-Distill-Qwen-1.5B: —, Phi-4: 37.2 (#237)
| Benchmark | DeepSeek-R1-Distill-Qwen-1.5B | Phi-4 |
|---|---|---|
| LMArena Non-English | — | 1197 |
| LMArena Chinese | — | 1212 |
| LMArena French | — | 1224 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1158 |
| LMArena Korean | — | 1151 |
| LMArena Russian | — | 1209 |
| LMArena Spanish | — | 1234 |
Instruction Following Not comparable
DeepSeek-R1-Distill-Qwen-1.5B: —, Phi-4: 60.4 (#251)
| Benchmark | DeepSeek-R1-Distill-Qwen-1.5B | Phi-4 |
|---|---|---|
| LiveBench Instruction Following | — | 58.4% |
| LMArena Instruction Following | — | 1201 |
Long Context Not comparable
DeepSeek-R1-Distill-Qwen-1.5B: —, Phi-4: 36.9 (#226)
| Benchmark | DeepSeek-R1-Distill-Qwen-1.5B | Phi-4 |
|---|---|---|
| LMArena Longer Query | — | 1217 |
Writing & Preference Not comparable
DeepSeek-R1-Distill-Qwen-1.5B: —, Phi-4: 40.5 (#244)
| Benchmark | DeepSeek-R1-Distill-Qwen-1.5B | Phi-4 |
|---|---|---|
| LMArena Text | — | 1217 |
| LMArena Creative Writing | — | 1182 |
| Short-Story Creative Writing | — | 62.6% |
| LMArena Multi-Turn | — | 1206 |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is DeepSeek-R1-Distill-Qwen-1.5B better than Phi-4?
Phi-4 is the stronger model overall, scoring 31.2 to 26.1 on the Noometry Index.
Is DeepSeek-R1-Distill-Qwen-1.5B or Phi-4 better for coding?
Phi-4 scores higher on coding benchmarks: 34.4 versus 21.8 in the Noometry coding category.
How many benchmarks do DeepSeek-R1-Distill-Qwen-1.5B and Phi-4 share?
5 benchmarks have published results for both models. DeepSeek-R1-Distill-Qwen-1.5B has 5 scored results on Noometry and Phi-4 has 37.