Model comparison
Mixtral 8x22B vs Step 3
Step 3 is the stronger model overall, scoring 40.5 to 27.1 on the Noometry Index.
Last verified . 15 shared benchmarks.
Mistral AI
StepFunSummary
- They share 15 benchmarks with published results for both. Mixtral 8x22B scores higher in 0 categories and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Step 3 leads 36.8 to 15.1.
Side by side
| Mixtral 8x22B | Step 3 | |
|---|---|---|
| Provider | Mistral AI | StepFun |
| Noometry Index | 27.1 | 40.5 |
| 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 Step 3 leads
Mixtral 8x22B: 24.2 (#329), Step 3: 40.1 (#147)
| Benchmark | Mixtral 8x22B | Step 3 |
|---|---|---|
| LMArena Coding | 1166 | 1367 |
| WeirdML | 3.2% | — |
| BigCodeBench Instruct | 40.6% | — |
| BigCodeBench Complete | 50.2% | — |
| HumanEval+ | 72% | — |
| MBPP+ | 64.3% | — |
Agentic & Tool Use Not comparable
Mixtral 8x22B: 23.1 (#127), Step 3: —
| Benchmark | Mixtral 8x22B | Step 3 |
|---|---|---|
| Cybench | 7.5% | — |
Reasoning Step 3 leads
Mixtral 8x22B: 19.9 (#248), Step 3: 28.4 (#105)
| Benchmark | Mixtral 8x22B | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1150 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| DTBench | 55.1% | — |
| Epoch Capabilities Index | 122.03 | — |
| ForecastBench | 56.3 | — |
Math Step 3 leads
Mixtral 8x22B: 22.9 (#275), Step 3: 37.6 (#148)
| Benchmark | Mixtral 8x22B | Step 3 |
|---|---|---|
| LMArena Math | 1184 | 1366 |
| Omni-MATH | 16.3% | — |
| MATH Level 5 | 24.2% | — |
Knowledge Step 3 leads
Mixtral 8x22B: 15.1 (#293), Step 3: 36.8 (#164)
| Benchmark | Mixtral 8x22B | Step 3 |
|---|---|---|
| LMArena Expert | 1113 | 1333 |
| GPQA Diamond | 34.1% | — |
| MMLU-Pro | 46% | — |
| GPQA (HELM) | 33.4% | — |
| MMLU | 77.8% | — |
Multimodal Not comparable
Mixtral 8x22B: —, Step 3: 35.5 (#86)
| Benchmark | Mixtral 8x22B | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Step 3 leads
Mixtral 8x22B: 32.8 (#255), Step 3: 46.3 (#159)
| Benchmark | Mixtral 8x22B | Step 3 |
|---|---|---|
| LMArena Non-English | 1128 | 1327 |
| LMArena Chinese | 1116 | 1397 |
| LMArena German | 1141 | 1371 |
| LMArena Korean | 1057 | 1269 |
| LMArena Russian | 1158 | 1331 |
| LMArena Spanish | 1151 | 1371 |
| LMArena French | 1166 | — |
| LMArena Japanese | 1037 | — |
Instruction Following Step 3 leads
Mixtral 8x22B: 57.7 (#266), Step 3: 70.4 (#164)
| Benchmark | Mixtral 8x22B | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1147 | 1332 |
| IFEval | 72.4% | — |
Long Context Step 3 leads
Mixtral 8x22B: 34.7 (#247), Step 3: 40.3 (#157)
| Benchmark | Mixtral 8x22B | Step 3 |
|---|---|---|
| LMArena Longer Query | 1144 | 1326 |
Writing & Preference Step 3 leads
Mixtral 8x22B: 36.9 (#262), Step 3: 54.3 (#151)
| Benchmark | Mixtral 8x22B | Step 3 |
|---|---|---|
| LMArena Text | 1162 | 1350 |
| LMArena Creative Writing | 1141 | 1321 |
| LMArena Multi-Turn | 1130 | 1341 |
| WildBench | 71.1% | — |
Frequently asked questions
Is Mixtral 8x22B better than Step 3?
Step 3 is the stronger model overall, scoring 40.5 to 27.1 on the Noometry Index.
Is Mixtral 8x22B or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 24.2 in the Noometry coding category.
How many benchmarks do Mixtral 8x22B and Step 3 share?
15 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and Step 3 has 17.