Model comparison
Ministral 8B vs Mixtral 8x22B
Ministral 8B is the stronger model overall, scoring 28.2 to 27.1 on the Noometry Index.
Last verified . 15 shared benchmarks.
Mistral AISummary
- They share 15 benchmarks with published results for both. Ministral 8B scores higher in 6 categories and Mixtral 8x22B in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where Ministral 8B leads 35.0 to 24.2.
- The biggest single-benchmark swing is DTBench: 45.7% for Ministral 8B and 55.1% for Mixtral 8x22B.
- Ministral 8B is cheaper at $0.15 / $0.15 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
- Ministral 8B accepts more context: 262K tokens versus 64K.
Side by side
| Ministral 8B | Mixtral 8x22B | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 28.2 | 27.1 |
| Released | 2024-10-01 | 2024-04-17 |
| Weights | Open | Open |
| Context window | 262K | 64K |
| Max output | 262K | 64K |
| Input $ / M tokens | $0.15 | $2 |
| Output $ / M tokens | $0.15 | $6 |
| Results tracked | 17 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Ministral 8B leads
Ministral 8B: 35.0 (#230), Mixtral 8x22B: 24.2 (#329)
| Benchmark | Ministral 8B | Mixtral 8x22B |
|---|---|---|
| LMArena Coding | 1202 | 1166 |
| WeirdML | — | 3.2% |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use Mixtral 8x22B leads
Ministral 8B: 16.4 (#148), Mixtral 8x22B: 23.1 (#127)
| Benchmark | Ministral 8B | Mixtral 8x22B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 11.1% | — |
| Cybench | — | 7.5% |
Reasoning Mixtral 8x22B leads
Ministral 8B: 18.4 (#281), Mixtral 8x22B: 19.9 (#248)
| Benchmark | Ministral 8B | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1150 |
| DTBench | 45.7% | 55.1% |
| Epoch Capabilities Index | — | 122.03 |
| ForecastBench | — | 56.3 |
Math Ministral 8B leads
Ministral 8B: 25.7 (#267), Mixtral 8x22B: 22.9 (#275)
| Benchmark | Ministral 8B | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 1188 | 1184 |
| MATH Level 5 | 14.9% | 24.2% |
| Omni-MATH | — | 16.3% |
Knowledge Mixtral 8x22B leads
Ministral 8B: 12.6 (#297), Mixtral 8x22B: 15.1 (#293)
| Benchmark | Ministral 8B | Mixtral 8x22B |
|---|---|---|
| GPQA Diamond | 27.1% | 34.1% |
| LMArena Expert | 1170 | 1113 |
| MMLU-Pro | — | 46% |
| Vectara Hallucination Rate | 7.4% | — |
| GPQA (HELM) | — | 33.4% |
| MMLU | — | 77.8% |
Multilingual Ministral 8B leads
Ministral 8B: 35.1 (#247), Mixtral 8x22B: 32.8 (#255)
| Benchmark | Ministral 8B | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1165 | 1128 |
| LMArena Chinese | 1193 | 1116 |
| LMArena Russian | 1195 | 1158 |
| LMArena French | — | 1166 |
| LMArena German | — | 1141 |
| LMArena Japanese | — | 1037 |
| LMArena Korean | — | 1057 |
| LMArena Spanish | — | 1151 |
Instruction Following Ministral 8B leads
Ministral 8B: 60.5 (#250), Mixtral 8x22B: 57.7 (#266)
| Benchmark | Ministral 8B | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 1161 | 1147 |
| IFEval | — | 72.4% |
Long Context Ministral 8B leads
Ministral 8B: 36.7 (#227), Mixtral 8x22B: 34.7 (#247)
| Benchmark | Ministral 8B | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1212 | 1144 |
Writing & Preference Ministral 8B leads
Ministral 8B: 39.6 (#246), Mixtral 8x22B: 36.9 (#262)
| Benchmark | Ministral 8B | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1191 | 1162 |
| LMArena Creative Writing | 1175 | 1141 |
| LMArena Multi-Turn | 1166 | 1130 |
| WildBench | — | 71.1% |
Frequently asked questions
Is Ministral 8B better than Mixtral 8x22B?
Ministral 8B is the stronger model overall, scoring 28.2 to 27.1 on the Noometry Index.
Which is cheaper, Ministral 8B or Mixtral 8x22B?
Ministral 8B is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; Mixtral 8x22B lists at $2 and $6.
Is Ministral 8B or Mixtral 8x22B better for coding?
Ministral 8B scores higher on coding benchmarks: 35.0 versus 24.2 in the Noometry coding category.
Which has the bigger context window?
Ministral 8B does, with 262K tokens against 64K.
How many benchmarks do Ministral 8B and Mixtral 8x22B share?
15 benchmarks have published results for both models. Ministral 8B has 17 scored results on Noometry and Mixtral 8x22B has 34.