Model comparison
Mixtral 8x22B vs Phi 3 Mini 4k Instruct
Mixtral 8x22B and Phi 3 Mini 4k Instruct score almost the same on the Noometry Index (27.1 vs 27.9), so choose on price, context window or the category you care about most.
Last verified . 20 shared benchmarks.
Mistral AI
MicrosoftSummary
- They share 20 benchmarks with published results for both. Mixtral 8x22B scores higher in 5 categories and Phi 3 Mini 4k Instruct in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Phi 3 Mini 4k Instruct leads 28.5 to 15.1.
Side by side
| Mixtral 8x22B | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | Mistral AI | Microsoft |
| Noometry Index | 27.1 | 27.9 |
| Released | 2024-04-17 | 2024-04-23 |
| Weights | Open | Open |
| Context window | 64K | — |
| Max output | 64K | — |
| Input $ / M tokens | $2 | — |
| Output $ / M tokens | $6 | — |
| Results tracked | 34 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi 3 Mini 4k Instruct leads
Mixtral 8x22B: 24.2 (#329), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | Mixtral 8x22B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1166 | 1093 |
| HumanEval+ | 72% | 59.1% |
| MBPP+ | 64.3% | 54.2% |
| WeirdML | 3.2% | — |
| BigCodeBench Instruct | 40.6% | — |
| LiveBench Coding | — | 15.5% |
| BigCodeBench Complete | 50.2% | — |
Agentic & Tool Use Not comparable
Mixtral 8x22B: 23.1 (#127), Phi 3 Mini 4k Instruct: —
| Benchmark | Mixtral 8x22B | Phi 3 Mini 4k Instruct |
|---|---|---|
| Cybench | 7.5% | — |
Reasoning Mixtral 8x22B leads
Mixtral 8x22B: 19.9 (#248), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | Mixtral 8x22B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1150 | 1072 |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 26.8% |
| DTBench | 55.1% | — |
| LiveBench Data Analysis | — | 34.7% |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 122.03 | — |
| ForecastBench | 56.3 | — |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |
Math Phi 3 Mini 4k Instruct leads
Mixtral 8x22B: 22.9 (#275), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | Mixtral 8x22B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1184 | 1111 |
| Omni-MATH | 16.3% | — |
| LiveBench Math | — | 15.7% |
| MATH Level 5 | 24.2% | — |
Knowledge Phi 3 Mini 4k Instruct leads
Mixtral 8x22B: 15.1 (#293), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | Mixtral 8x22B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1113 | 1045 |
| MMLU | 77.8% | 68.8% |
| GPQA Diamond | 34.1% | — |
| MMLU-Pro | 46% | — |
| GPQA (HELM) | 33.4% | — |
| ARC (AI2) Challenge | — | 84.9% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multilingual Mixtral 8x22B leads
Mixtral 8x22B: 32.8 (#255), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | Mixtral 8x22B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1128 | 1021 |
| LMArena Chinese | 1116 | 1021 |
| LMArena French | 1166 | 1076 |
| LMArena German | 1141 | 1044 |
| LMArena Japanese | 1037 | 935 |
| LMArena Korean | 1057 | 905 |
| LMArena Russian | 1158 | 1022 |
| LMArena Spanish | 1151 | 1085 |
Instruction Following Mixtral 8x22B leads
Mixtral 8x22B: 57.7 (#266), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | Mixtral 8x22B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1147 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
| IFEval | 72.4% | — |
Long Context Mixtral 8x22B leads
Mixtral 8x22B: 34.7 (#247), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | Mixtral 8x22B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1144 | 1044 |
Writing & Preference Mixtral 8x22B leads
Mixtral 8x22B: 36.9 (#262), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | Mixtral 8x22B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1162 | 1073 |
| LMArena Creative Writing | 1141 | 1037 |
| LMArena Multi-Turn | 1130 | 1018 |
| WildBench | 71.1% | — |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is Mixtral 8x22B better than Phi 3 Mini 4k Instruct?
Mixtral 8x22B and Phi 3 Mini 4k Instruct score almost the same on the Noometry Index (27.1 vs 27.9), so choose on price, context window or the category you care about most.
Is Mixtral 8x22B or Phi 3 Mini 4k Instruct better for coding?
Phi 3 Mini 4k Instruct scores higher on coding benchmarks: 26.6 versus 24.2 in the Noometry coding category.
How many benchmarks do Mixtral 8x22B and Phi 3 Mini 4k Instruct share?
20 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.