Model comparison
Phi 3 Mini 4k Instruct vs Qwen2-72B
Qwen2-72B is the stronger model overall, scoring 30.0 to 27.9 on the Noometry Index.
Last verified . 18 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 18 benchmarks with published results for both. Phi 3 Mini 4k Instruct scores higher in 1 category and Qwen2-72B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Qwen2-72B leads 61.7 to 47.7.
Side by side
| Phi 3 Mini 4k Instruct | Qwen2-72B | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 27.9 | 30.0 |
| Released | 2024-04-23 | 2024-06-07 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2-72B leads
Phi 3 Mini 4k Instruct: 26.6 (#323), Qwen2-72B: 29.1 (#310)
| Benchmark | Phi 3 Mini 4k Instruct | Qwen2-72B |
|---|---|---|
| LMArena Coding | 1093 | 1196 |
| WeirdML | — | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| LiveBench Coding | 15.5% | — |
| BigCodeBench Complete | — | 54% |
| HumanEval+ | 59.1% | — |
| MBPP+ | 54.2% | — |
Agentic & Tool Use Not comparable
Phi 3 Mini 4k Instruct: —, Qwen2-72B: 17.0 (#146)
| Benchmark | Phi 3 Mini 4k Instruct | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
Phi 3 Mini 4k Instruct: 14.1 (#328), Qwen2-72B: 23.2 (#181)
| Benchmark | Phi 3 Mini 4k Instruct | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1072 | 1191 |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 26.8% | — |
| LiveBench Data Analysis | 34.7% | — |
| Adversarial NLI | 52.8% | — |
| BIG-Bench Hard | 71.7% | — |
| Epoch Capabilities Index | — | 125.28 |
| HellaSwag | 76.7% | — |
| LiveBench | 22.4% | — |
| WinoGrande | 70.8% | — |
Math Qwen2-72B leads
Phi 3 Mini 4k Instruct: 26.6 (#257), Qwen2-72B: 30.2 (#236)
| Benchmark | Phi 3 Mini 4k Instruct | Qwen2-72B |
|---|---|---|
| LMArena Math | 1111 | 1235 |
| LiveBench Math | 15.7% | — |
| MATH Level 5 | — | 39.1% |
Knowledge Phi 3 Mini 4k Instruct leads
Phi 3 Mini 4k Instruct: 28.5 (#246), Qwen2-72B: 21.2 (#275)
| Benchmark | Phi 3 Mini 4k Instruct | Qwen2-72B |
|---|---|---|
| LMArena Expert | 1045 | 1171 |
| MMLU | 68.8% | 82.4% |
| GPQA Diamond | — | 40.8% |
| ARC (AI2) Challenge | 84.9% | — |
| OpenBookQA | 88% | — |
| TriviaQA | 64% | — |
Multilingual Qwen2-72B leads
Phi 3 Mini 4k Instruct: 26.3 (#280), Qwen2-72B: 35.9 (#244)
| Benchmark | Phi 3 Mini 4k Instruct | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1021 | 1176 |
| LMArena Chinese | 1021 | 1240 |
| LMArena French | 1076 | 1170 |
| LMArena German | 1044 | 1151 |
| LMArena Japanese | 935 | 1111 |
| LMArena Korean | 905 | 1083 |
| LMArena Russian | 1022 | 1169 |
| LMArena Spanish | 1085 | 1169 |
Instruction Following Qwen2-72B leads
Phi 3 Mini 4k Instruct: 47.7 (#303), Qwen2-72B: 61.7 (#241)
| Benchmark | Phi 3 Mini 4k Instruct | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1053 | 1181 |
| LiveBench Instruction Following | 39.1% | — |
Long Context Qwen2-72B leads
Phi 3 Mini 4k Instruct: 31.7 (#276), Qwen2-72B: 36.1 (#235)
| Benchmark | Phi 3 Mini 4k Instruct | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1044 | 1192 |
Writing & Preference Qwen2-72B leads
Phi 3 Mini 4k Instruct: 27.6 (#300), Qwen2-72B: 40.8 (#241)
| Benchmark | Phi 3 Mini 4k Instruct | Qwen2-72B |
|---|---|---|
| LMArena Text | 1073 | 1203 |
| LMArena Creative Writing | 1037 | 1181 |
| LMArena Multi-Turn | 1018 | 1196 |
| LiveBench Language | 9.2% | — |
Frequently asked questions
Is Phi 3 Mini 4k Instruct better than Qwen2-72B?
Qwen2-72B is the stronger model overall, scoring 30.0 to 27.9 on the Noometry Index.
Is Phi 3 Mini 4k Instruct or Qwen2-72B better for coding?
Qwen2-72B scores higher on coding benchmarks: 29.1 versus 26.6 in the Noometry coding category.
How many benchmarks do Phi 3 Mini 4k Instruct and Qwen2-72B share?
18 benchmarks have published results for both models. Phi 3 Mini 4k Instruct has 35 scored results on Noometry and Qwen2-72B has 26.