Model comparison
Llama 3-8B vs Phi 3 Mini 4k Instruct
Phi 3 Mini 4k Instruct is the stronger model overall, scoring 27.4 to 25.3 on the Noometry Index.
Last verified . 26 shared benchmarks.
Meta
MicrosoftSummary
- They share 26 benchmarks with published results for both. Llama 3-8B scores higher in 6 categories and Phi 3 Mini 4k Instruct in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Phi 3 Mini 4k Instruct leads 27.9 to 7.8.
Side by side
| Llama 3-8B | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | Meta | Microsoft |
| Noometry Index | 25.3 | 27.4 |
| Released | 2024-04-18 | 2024-04-23 |
| Weights | Open | Open |
| Context window | 8K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.14 | — |
| Output $ / M tokens | $0.14 | — |
| Results tracked | 34 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-8B leads
Llama 3-8B: 30.5 (#285), Phi 3 Mini 4k Instruct: 26.1 (#319)
| Benchmark | Llama 3-8B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1152 | 1093 |
| HumanEval+ | 56.7% | 59.1% |
| MBPP+ | 54.8% | 54.2% |
| BigCodeBench Instruct | 31.9% | — |
| LiveBench Coding | — | 15.5% |
| BigCodeBench Complete | 36.9% | — |
Reasoning Too close to call
Llama 3-8B: 13.8 (#321), Phi 3 Mini 4k Instruct: 13.6 (#326)
| Benchmark | Llama 3-8B | Phi 3 Mini 4k Instruct |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1133 | 1072 |
| Adversarial NLI | 57.3% | 52.8% |
| WinoGrande | 75.7% | 70.8% |
| LiveBench Reasoning | — | 26.8% |
| DTBench | 43.9% | — |
| LiveBench Data Analysis | — | 34.7% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 116.33 | — |
| ForecastBench | 58.6 | — |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
Math Phi 3 Mini 4k Instruct leads
Llama 3-8B: 8.3 (#317), Phi 3 Mini 4k Instruct: 24.5 (#258)
| Benchmark | Llama 3-8B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1151 | 1111 |
| OTIS Mock AIME 2024-2025 | 1.9% | — |
| LiveBench Math | — | 15.7% |
| MATH Level 5 | 6.1% | — |
Knowledge Phi 3 Mini 4k Instruct leads
Llama 3-8B: 7.8 (#303), Phi 3 Mini 4k Instruct: 27.9 (#241)
| Benchmark | Llama 3-8B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1113 | 1045 |
| ARC (AI2) Challenge | 82.8% | 84.9% |
| MMLU | 68.8% | 68.8% |
| OpenBookQA | 82.6% | 88% |
| TriviaQA | 67.7% | 64% |
| GPQA Diamond | 26.1% | — |
Multilingual Llama 3-8B leads
Llama 3-8B: 31.0 (#256), Phi 3 Mini 4k Instruct: 26.5 (#275)
| Benchmark | Llama 3-8B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1098 | 1021 |
| LMArena Chinese | 1076 | 1021 |
| LMArena French | 1158 | 1075 |
| LMArena German | 1104 | 1044 |
| LMArena Japanese | 967 | 935 |
| LMArena Korean | 1003 | 905 |
| LMArena Russian | 1109 | 1022 |
| LMArena Spanish | 1172 | 1084 |
Instruction Following Llama 3-8B leads
Llama 3-8B: 58.7 (#255), Phi 3 Mini 4k Instruct: 48.0 (#298)
| Benchmark | Llama 3-8B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1127 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
Long Context Llama 3-8B leads
Llama 3-8B: 34.2 (#246), Phi 3 Mini 4k Instruct: 31.7 (#271)
| Benchmark | Llama 3-8B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1128 | 1044 |
Writing & Preference Llama 3-8B leads
Llama 3-8B: 37.7 (#251), Phi 3 Mini 4k Instruct: 27.7 (#295)
| Benchmark | Llama 3-8B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1166 | 1073 |
| LMArena Creative Writing | 1150 | 1037 |
| LMArena Multi-Turn | 1152 | 1018 |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is Llama 3-8B better than Phi 3 Mini 4k Instruct?
Phi 3 Mini 4k Instruct is the stronger model overall, scoring 27.4 to 25.3 on the Noometry Index.
Is Llama 3-8B or Phi 3 Mini 4k Instruct better for coding?
Llama 3-8B scores higher on coding benchmarks: 30.5 versus 26.1 in the Noometry coding category.
How many benchmarks do Llama 3-8B and Phi 3 Mini 4k Instruct share?
26 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.