Model comparison
Phi 3 Mini 4k Instruct vs Wizardlm 70b
Wizardlm 70b is the stronger model overall, scoring 33.0 to 27.9 on the Noometry Index.
Last verified . 12 shared benchmarks.
MicrosoftSummary
- They share 12 benchmarks with published results for both. Phi 3 Mini 4k Instruct scores higher in 0 categories and Wizardlm 70b in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Wizardlm 70b leads 56.3 to 47.7.
Side by side
| Phi 3 Mini 4k Instruct | Wizardlm 70b | |
|---|---|---|
| Provider | Microsoft | Microsoft |
| Noometry Index | 27.9 | 33.0 |
| Released | 2024-04-23 | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Wizardlm 70b leads
Phi 3 Mini 4k Instruct: 26.6 (#323), Wizardlm 70b: 31.4 (#285)
| Benchmark | Phi 3 Mini 4k Instruct | Wizardlm 70b |
|---|---|---|
| LMArena Coding | 1093 | 1081 |
| LiveBench Coding | 15.5% | — |
| HumanEval+ | 59.1% | — |
| MBPP+ | 54.2% | — |
Reasoning Wizardlm 70b leads
Phi 3 Mini 4k Instruct: 14.1 (#328), Wizardlm 70b: 20.7 (#234)
| Benchmark | Phi 3 Mini 4k Instruct | Wizardlm 70b |
|---|---|---|
| LMArena Hard Prompts | 1072 | 1079 |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 26.8% | — |
| LiveBench Data Analysis | 34.7% | — |
| Adversarial NLI | 52.8% | — |
| BIG-Bench Hard | 71.7% | — |
| HellaSwag | 76.7% | — |
| LiveBench | 22.4% | — |
| WinoGrande | 70.8% | — |
Math Wizardlm 70b leads
Phi 3 Mini 4k Instruct: 26.6 (#257), Wizardlm 70b: 32.2 (#218)
| Benchmark | Phi 3 Mini 4k Instruct | Wizardlm 70b |
|---|---|---|
| LMArena Math | 1111 | 1116 |
| LiveBench Math | 15.7% | — |
Knowledge Not comparable
Phi 3 Mini 4k Instruct: 28.5 (#246), Wizardlm 70b: —
| Benchmark | Phi 3 Mini 4k Instruct | Wizardlm 70b |
|---|---|---|
| LMArena Expert | 1045 | — |
| ARC (AI2) Challenge | 84.9% | — |
| MMLU | 68.8% | — |
| OpenBookQA | 88% | — |
| TriviaQA | 64% | — |
Multilingual Wizardlm 70b leads
Phi 3 Mini 4k Instruct: 26.3 (#280), Wizardlm 70b: 29.6 (#265)
| Benchmark | Phi 3 Mini 4k Instruct | Wizardlm 70b |
|---|---|---|
| LMArena Non-English | 1021 | 1078 |
| LMArena Chinese | 1021 | 1052 |
| LMArena German | 1044 | 1083 |
| LMArena Russian | 1022 | 1155 |
| LMArena French | 1076 | — |
| LMArena Japanese | 935 | — |
| LMArena Korean | 905 | — |
| LMArena Spanish | 1085 | — |
Instruction Following Wizardlm 70b leads
Phi 3 Mini 4k Instruct: 47.7 (#303), Wizardlm 70b: 56.3 (#273)
| Benchmark | Phi 3 Mini 4k Instruct | Wizardlm 70b |
|---|---|---|
| LMArena Instruction Following | 1053 | 1093 |
| LiveBench Instruction Following | 39.1% | — |
Long Context Wizardlm 70b leads
Phi 3 Mini 4k Instruct: 31.7 (#276), Wizardlm 70b: 33.3 (#263)
| Benchmark | Phi 3 Mini 4k Instruct | Wizardlm 70b |
|---|---|---|
| LMArena Longer Query | 1044 | 1097 |
Writing & Preference Wizardlm 70b leads
Phi 3 Mini 4k Instruct: 27.6 (#300), Wizardlm 70b: 34.8 (#269)
| Benchmark | Phi 3 Mini 4k Instruct | Wizardlm 70b |
|---|---|---|
| LMArena Text | 1073 | 1120 |
| LMArena Creative Writing | 1037 | 1149 |
| LMArena Multi-Turn | 1018 | 1108 |
| LiveBench Language | 9.2% | — |
Frequently asked questions
Is Phi 3 Mini 4k Instruct better than Wizardlm 70b?
Wizardlm 70b is the stronger model overall, scoring 33.0 to 27.9 on the Noometry Index.
Is Phi 3 Mini 4k Instruct or Wizardlm 70b better for coding?
Wizardlm 70b scores higher on coding benchmarks: 31.4 versus 26.6 in the Noometry coding category.
How many benchmarks do Phi 3 Mini 4k Instruct and Wizardlm 70b share?
12 benchmarks have published results for both models. Phi 3 Mini 4k Instruct has 35 scored results on Noometry and Wizardlm 70b has 12.