Model comparison
Mixtral 8x22B vs Phi 3 Medium 4k Instruct
Phi 3 Medium 4k Instruct is the stronger model overall, scoring 32.4 to 26.7 on the Noometry Index.
Last verified . 17 shared benchmarks.
Mistral AI
MicrosoftSummary
- They share 17 benchmarks with published results for both. Mixtral 8x22B scores higher in 4 categories and Phi 3 Medium 4k Instruct in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Phi 3 Medium 4k Instruct leads 29.6 to 15.1.
Side by side
| Mixtral 8x22B | Phi 3 Medium 4k Instruct | |
|---|---|---|
| Provider | Mistral AI | Microsoft |
| Noometry Index | 26.7 | 32.4 |
| Released | 2024-04-17 | — |
| Weights | Open | Open |
| Context window | 64K | — |
| Max output | 64K | — |
| Input $ / M tokens | $2 | — |
| Output $ / M tokens | $6 | — |
| Results tracked | 34 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi 3 Medium 4k Instruct leads
Mixtral 8x22B: 23.8 (#324), Phi 3 Medium 4k Instruct: 32.3 (#263)
| Benchmark | Mixtral 8x22B | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Coding | 1166 | 1130 |
| WeirdML | 3.2% | — |
| BigCodeBench Instruct | 40.6% | — |
| BigCodeBench Complete | 50.2% | — |
| HumanEval+ | 72% | — |
| MBPP+ | 64.3% | — |
Agentic & Tool Use Not comparable
Mixtral 8x22B: 22.8 (#119), Phi 3 Medium 4k Instruct: —
| Benchmark | Mixtral 8x22B | Phi 3 Medium 4k Instruct |
|---|---|---|
| Cybench | 7.5% | — |
Reasoning Phi 3 Medium 4k Instruct leads
Mixtral 8x22B: 19.1 (#246), Phi 3 Medium 4k Instruct: 20.8 (#214)
| Benchmark | Mixtral 8x22B | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1150 | 1127 |
| DTBench | 55.1% | — |
| Epoch Capabilities Index | 122 | — |
| ForecastBench | 56.3 | — |
Math Phi 3 Medium 4k Instruct leads
Mixtral 8x22B: 21.2 (#271), Phi 3 Medium 4k Instruct: 30.9 (#202)
| Benchmark | Mixtral 8x22B | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Math | 1184 | 1173 |
| Omni-MATH | 16.3% | — |
| MATH Level 5 | 24.2% | — |
Knowledge Phi 3 Medium 4k Instruct leads
Mixtral 8x22B: 15.1 (#288), Phi 3 Medium 4k Instruct: 29.6 (#226)
| Benchmark | Mixtral 8x22B | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Expert | 1113 | 1108 |
| GPQA Diamond | 34.1% | — |
| MMLU-Pro | 46% | — |
| GPQA (HELM) | 33.4% | — |
| MMLU | 77.8% | — |
Multilingual Mixtral 8x22B leads
Mixtral 8x22B: 32.9 (#250), Phi 3 Medium 4k Instruct: 30.8 (#258)
| Benchmark | Mixtral 8x22B | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Non-English | 1128 | 1094 |
| LMArena Chinese | 1116 | 1108 |
| LMArena French | 1165 | 1109 |
| LMArena German | 1141 | 1101 |
| LMArena Japanese | 1037 | 1042 |
| LMArena Korean | 1056 | 954 |
| LMArena Russian | 1158 | 1145 |
| LMArena Spanish | 1150 | 1095 |
Instruction Following Too close to call
Mixtral 8x22B: 57.9 (#261), Phi 3 Medium 4k Instruct: 57.8 (#263)
| Benchmark | Mixtral 8x22B | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1147 | 1114 |
| IFEval | 72.4% | — |
Long Context Too close to call
Mixtral 8x22B: 34.6 (#242), Phi 3 Medium 4k Instruct: 33.9 (#250)
| Benchmark | Mixtral 8x22B | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1144 | 1120 |
Writing & Preference Mixtral 8x22B leads
Mixtral 8x22B: 37.1 (#257), Phi 3 Medium 4k Instruct: 34.3 (#267)
| Benchmark | Mixtral 8x22B | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Text | 1162 | 1138 |
| LMArena Creative Writing | 1141 | 1106 |
| LMArena Multi-Turn | 1130 | 1088 |
| WildBench | 71.1% | — |
Frequently asked questions
Is Mixtral 8x22B better than Phi 3 Medium 4k Instruct?
Phi 3 Medium 4k Instruct is the stronger model overall, scoring 32.4 to 26.7 on the Noometry Index.
Is Mixtral 8x22B or Phi 3 Medium 4k Instruct better for coding?
Phi 3 Medium 4k Instruct scores higher on coding benchmarks: 32.3 versus 23.8 in the Noometry coding category.
How many benchmarks do Mixtral 8x22B and Phi 3 Medium 4k Instruct share?
17 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and Phi 3 Medium 4k Instruct has 17.