Model comparison
Llama 3.1-405B vs Phi 3 Mini 4k Instruct
Llama 3.1-405B is the stronger model overall, scoring 30.7 to 27.9 on the Noometry Index.
Last verified . 23 shared benchmarks.
Meta
MicrosoftSummary
- They share 23 benchmarks with published results for both. Llama 3.1-405B scores higher in 7 categories and Phi 3 Mini 4k Instruct in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Llama 3.1-405B leads 65.9 to 47.7.
Side by side
| Llama 3.1-405B | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | Meta | Microsoft |
| Noometry Index | 30.7 | 27.9 |
| Released | 2024-07-23 | 2024-04-23 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 42 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3.1-405B leads
Llama 3.1-405B: 33.1 (#262), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | Llama 3.1-405B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1291 | 1093 |
| WeirdML | 21.4% | — |
| LiveBench Coding | — | 15.5% |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Agentic & Tool Use Not comparable
Llama 3.1-405B: 21.0 (#140), Phi 3 Mini 4k Instruct: —
| Benchmark | Llama 3.1-405B | Phi 3 Mini 4k Instruct |
|---|---|---|
| TheAgentCompany | 7.4% | — |
| Cybench | 7.5% | — |
Reasoning Llama 3.1-405B leads
Llama 3.1-405B: 16.8 (#300), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | Llama 3.1-405B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1269 | 1072 |
| BIG-Bench Hard | 82.9% | 71.7% |
| HellaSwag | 89.2% | 76.7% |
| WinoGrande | 89.2% | 70.8% |
| SimpleBench | 23% | — |
| Kagi LLM Benchmark | 45% | — |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 26.8% |
| DTBench | 61.4% | — |
| LiveBench Data Analysis | — | 34.7% |
| Adversarial NLI | — | 52.8% |
| Epoch Capabilities Index | 128.75 | — |
| ForecastBench | 59.9 | — |
| LiveBench | — | 22.4% |
| PIQA | 85.9% | — |
Math Phi 3 Mini 4k Instruct leads
Llama 3.1-405B: 18.4 (#290), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | Llama 3.1-405B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1281 | 1111 |
| OTIS Mock AIME 2024-2025 | 9.7% | — |
| Omni-MATH | 24.9% | — |
| LiveBench Math | — | 15.7% |
| MATH Level 5 | 49.8% | — |
Knowledge Llama 3.1-405B leads
Llama 3.1-405B: 30.4 (#227), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | Llama 3.1-405B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1243 | 1045 |
| ARC (AI2) Challenge | 95.3% | 84.9% |
| MMLU | 84.5% | 68.8% |
| TriviaQA | 82.7% | 64% |
| GPQA Diamond | 50.9% | — |
| MMLU-Pro | 72.3% | — |
| Confabulations | 17.6% | — |
| GPQA (HELM) | 52.2% | — |
| OpenBookQA | — | 88% |
Multilingual Llama 3.1-405B leads
Llama 3.1-405B: 40.7 (#214), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | Llama 3.1-405B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1248 | 1021 |
| LMArena Chinese | 1242 | 1021 |
| LMArena French | 1279 | 1076 |
| LMArena German | 1252 | 1044 |
| LMArena Japanese | 1208 | 935 |
| LMArena Korean | 1184 | 905 |
| LMArena Russian | 1265 | 1022 |
| LMArena Spanish | 1260 | 1085 |
Instruction Following Llama 3.1-405B leads
Llama 3.1-405B: 65.9 (#214), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | Llama 3.1-405B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1259 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
| IFEval | 81.1% | — |
Long Context Llama 3.1-405B leads
Llama 3.1-405B: 38.4 (#197), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | Llama 3.1-405B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1266 | 1044 |
Writing & Preference Llama 3.1-405B leads
Llama 3.1-405B: 38.9 (#251), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | Llama 3.1-405B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1284 | 1073 |
| LMArena Creative Writing | 1262 | 1037 |
| LMArena Multi-Turn | 1297 | 1018 |
| EQ-Bench Creative Writing | 870 | — |
| WildBench | 78.3% | — |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is Llama 3.1-405B better than Phi 3 Mini 4k Instruct?
Llama 3.1-405B is the stronger model overall, scoring 30.7 to 27.9 on the Noometry Index.
Is Llama 3.1-405B or Phi 3 Mini 4k Instruct better for coding?
Llama 3.1-405B scores higher on coding benchmarks: 33.1 versus 26.6 in the Noometry coding category.
How many benchmarks do Llama 3.1-405B and Phi 3 Mini 4k Instruct share?
23 benchmarks have published results for both models. Llama 3.1-405B has 42 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.