Model comparison
Phi 3 Mini 4k Instruct vs Phi 3 Small 8k Instruct
Phi 3 Small 8k Instruct is the stronger model overall, scoring 29.3 to 27.9 on the Noometry Index.
Last verified . 32 shared benchmarks.
MicrosoftSummary
- They share 32 benchmarks with published results for both. Phi 3 Mini 4k Instruct scores higher in 0 categories and Phi 3 Small 8k Instruct in 8 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Phi 3 Small 8k Instruct leads 51.9 to 47.7.
- The biggest single-benchmark swing is LiveBench Reasoning: 26.8% for Phi 3 Mini 4k Instruct and 15.9% for Phi 3 Small 8k Instruct.
Side by side
| Phi 3 Mini 4k Instruct | Phi 3 Small 8k Instruct | |
|---|---|---|
| Provider | Microsoft | Microsoft |
| Noometry Index | 27.9 | 29.3 |
| Released | 2024-04-23 | 2024-04-23 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi 3 Small 8k Instruct leads
Phi 3 Mini 4k Instruct: 26.6 (#323), Phi 3 Small 8k Instruct: 27.9 (#318)
| Benchmark | Phi 3 Mini 4k Instruct | Phi 3 Small 8k Instruct |
|---|---|---|
| LiveBench Coding | 15.5% | 20.3% |
| LMArena Coding | 1093 | 1101 |
| HumanEval+ | 59.1% | — |
| MBPP+ | 54.2% | — |
Reasoning Too close to call
Phi 3 Mini 4k Instruct: 14.1 (#328), Phi 3 Small 8k Instruct: 14.8 (#323)
| Benchmark | Phi 3 Mini 4k Instruct | Phi 3 Small 8k Instruct |
|---|---|---|
| LiveBench Reasoning | 26.8% | 15.9% |
| LMArena Hard Prompts | 1072 | 1100 |
| LiveBench Data Analysis | 34.7% | 30.3% |
| Adversarial NLI | 52.8% | 58.1% |
| BIG-Bench Hard | 71.7% | 79.1% |
| HellaSwag | 76.7% | 77% |
| LiveBench | 22.4% | 24% |
| WinoGrande | 70.8% | 81.5% |
| Chess Puzzles | 0% | — |
Math Too close to call
Phi 3 Mini 4k Instruct: 26.6 (#257), Phi 3 Small 8k Instruct: 27.6 (#248)
| Benchmark | Phi 3 Mini 4k Instruct | Phi 3 Small 8k Instruct |
|---|---|---|
| LiveBench Math | 15.7% | 17.6% |
| LMArena Math | 1111 | 1151 |
Knowledge Too close to call
Phi 3 Mini 4k Instruct: 28.5 (#246), Phi 3 Small 8k Instruct: 29.1 (#240)
| Benchmark | Phi 3 Mini 4k Instruct | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Expert | 1045 | 1067 |
| ARC (AI2) Challenge | 84.9% | 90.7% |
| MMLU | 68.8% | 75.7% |
| OpenBookQA | 88% | 88% |
| TriviaQA | 64% | 58.1% |
Multilingual Phi 3 Small 8k Instruct leads
Phi 3 Mini 4k Instruct: 26.3 (#280), Phi 3 Small 8k Instruct: 28.5 (#272)
| Benchmark | Phi 3 Mini 4k Instruct | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Non-English | 1021 | 1058 |
| LMArena Chinese | 1021 | 1061 |
| LMArena French | 1076 | 1135 |
| LMArena German | 1044 | 1080 |
| LMArena Japanese | 935 | 966 |
| LMArena Korean | 905 | 894 |
| LMArena Russian | 1022 | 1111 |
| LMArena Spanish | 1085 | 1111 |
Instruction Following Phi 3 Small 8k Instruct leads
Phi 3 Mini 4k Instruct: 47.7 (#303), Phi 3 Small 8k Instruct: 51.9 (#292)
| Benchmark | Phi 3 Mini 4k Instruct | Phi 3 Small 8k Instruct |
|---|---|---|
| LiveBench Instruction Following | 39.1% | 47.2% |
| LMArena Instruction Following | 1053 | 1087 |
Long Context Phi 3 Small 8k Instruct leads
Phi 3 Mini 4k Instruct: 31.7 (#276), Phi 3 Small 8k Instruct: 33.0 (#267)
| Benchmark | Phi 3 Mini 4k Instruct | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Longer Query | 1044 | 1088 |
Writing & Preference Phi 3 Small 8k Instruct leads
Phi 3 Mini 4k Instruct: 27.6 (#300), Phi 3 Small 8k Instruct: 31.1 (#284)
| Benchmark | Phi 3 Mini 4k Instruct | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Text | 1073 | 1110 |
| LMArena Creative Writing | 1037 | 1083 |
| LMArena Multi-Turn | 1018 | 1068 |
| LiveBench Language | 9.2% | 12.9% |
Frequently asked questions
Is Phi 3 Mini 4k Instruct better than Phi 3 Small 8k Instruct?
Phi 3 Small 8k Instruct is the stronger model overall, scoring 29.3 to 27.9 on the Noometry Index.
Is Phi 3 Mini 4k Instruct or Phi 3 Small 8k Instruct better for coding?
Phi 3 Small 8k Instruct scores higher on coding benchmarks: 27.9 versus 26.6 in the Noometry coding category.
How many benchmarks do Phi 3 Mini 4k Instruct and Phi 3 Small 8k Instruct share?
32 benchmarks have published results for both models. Phi 3 Mini 4k Instruct has 35 scored results on Noometry and Phi 3 Small 8k Instruct has 32.