Model comparison
Llama 3-8B vs phi-3-medium 14B
phi-3-medium 14B is the stronger model overall, scoring 29.1 to 25.3 on the Noometry Index.
Last verified . 11 shared benchmarks.
Meta
MicrosoftSummary
- They share 11 benchmarks with published results for both. Llama 3-8B scores higher in 0 categories and phi-3-medium 14B in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where phi-3-medium 14B leads 25.2 to 8.3.
- The biggest single-benchmark swing is BigCodeBench Complete: 36.9% for Llama 3-8B and 48.7% for phi-3-medium 14B.
Side by side
| Llama 3-8B | phi-3-medium 14B | |
|---|---|---|
| Provider | Meta | Microsoft |
| Noometry Index | 25.3 | 29.1 |
| 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 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding phi-3-medium 14B leads
Llama 3-8B: 30.5 (#285), phi-3-medium 14B: 36.3 (#198)
| Benchmark | Llama 3-8B | phi-3-medium 14B |
|---|---|---|
| BigCodeBench Instruct | 31.9% | 37.6% |
| BigCodeBench Complete | 36.9% | 48.7% |
| LMArena Coding | 1152 | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |
Reasoning Not comparable
Llama 3-8B: 13.8 (#321), phi-3-medium 14B: —
| Benchmark | Llama 3-8B | phi-3-medium 14B |
|---|---|---|
| Adversarial NLI | 57.3% | 55.8% |
| Epoch Capabilities Index | 116.33 | 121.18 |
| WinoGrande | 75.7% | 81.5% |
| Chess Puzzles | 0% | — |
| LMArena Hard Prompts | 1133 | — |
| DTBench | 43.9% | — |
| BIG-Bench Hard | — | 81.4% |
| ForecastBench | 58.6 | — |
| HellaSwag | — | 82.4% |
Math phi-3-medium 14B leads
Llama 3-8B: 8.3 (#317), phi-3-medium 14B: 25.2 (#249)
| Benchmark | Llama 3-8B | phi-3-medium 14B |
|---|---|---|
| MATH Level 5 | 6.1% | 17.6% |
| OTIS Mock AIME 2024-2025 | 1.9% | — |
| LMArena Math | 1151 | — |
Knowledge phi-3-medium 14B leads
Llama 3-8B: 7.8 (#303), phi-3-medium 14B: 9.1 (#301)
| Benchmark | Llama 3-8B | phi-3-medium 14B |
|---|---|---|
| GPQA Diamond | 26.1% | 27.6% |
| ARC (AI2) Challenge | 82.8% | 91.6% |
| MMLU | 68.8% | 78% |
| OpenBookQA | 82.6% | 87.4% |
| TriviaQA | 67.7% | 73.9% |
| LMArena Expert | 1113 | — |
Multilingual Not comparable
Llama 3-8B: 31.0 (#256), phi-3-medium 14B: —
| Benchmark | Llama 3-8B | phi-3-medium 14B |
|---|---|---|
| LMArena Non-English | 1098 | — |
| LMArena Chinese | 1076 | — |
| LMArena French | 1158 | — |
| LMArena German | 1104 | — |
| LMArena Japanese | 967 | — |
| LMArena Korean | 1003 | — |
| LMArena Russian | 1109 | — |
| LMArena Spanish | 1172 | — |
Instruction Following Not comparable
Llama 3-8B: 58.7 (#255), phi-3-medium 14B: —
| Benchmark | Llama 3-8B | phi-3-medium 14B |
|---|---|---|
| LMArena Instruction Following | 1127 | — |
Long Context Not comparable
Llama 3-8B: 34.2 (#246), phi-3-medium 14B: —
| Benchmark | Llama 3-8B | phi-3-medium 14B |
|---|---|---|
| LMArena Longer Query | 1128 | — |
Writing & Preference Not comparable
Llama 3-8B: 37.7 (#251), phi-3-medium 14B: —
| Benchmark | Llama 3-8B | phi-3-medium 14B |
|---|---|---|
| LMArena Text | 1166 | — |
| LMArena Creative Writing | 1150 | — |
| LMArena Multi-Turn | 1152 | — |
Frequently asked questions
Is Llama 3-8B better than phi-3-medium 14B?
phi-3-medium 14B is the stronger model overall, scoring 29.1 to 25.3 on the Noometry Index.
Is Llama 3-8B or phi-3-medium 14B better for coding?
phi-3-medium 14B scores higher on coding benchmarks: 36.3 versus 30.5 in the Noometry coding category.
How many benchmarks do Llama 3-8B and phi-3-medium 14B share?
11 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and phi-3-medium 14B has 13.