Model comparison
Llama 13b vs Mixtral 8x22B
Mixtral 8x22B is the stronger model overall, scoring 27.1 to 24.4 on the Noometry Index.
Last verified . 10 shared benchmarks.
Meta
Mistral AISummary
- They share 10 benchmarks with published results for both. Llama 13b scores higher in 1 category and Mixtral 8x22B in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mixtral 8x22B leads 36.9 to 13.8.
Side by side
| Llama 13b | Mixtral 8x22B | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 24.4 | 27.1 |
| Released | 2023-02-24 | 2024-04-17 |
| Weights | Open | Open |
| Context window | — | 64K |
| Max output | — | 64K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $6 |
| Results tracked | 21 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mixtral 8x22B leads
Llama 13b: 21.4 (#337), Mixtral 8x22B: 24.2 (#329)
| Benchmark | Llama 13b | Mixtral 8x22B |
|---|---|---|
| LMArena Coding | 683 | 1166 |
| WeirdML | — | 3.2% |
| BigCodeBench Instruct | — | 40.6% |
| BigCodeBench Complete | — | 50.2% |
| HumanEval+ | — | 72% |
| MBPP+ | — | 64.3% |
Agentic & Tool Use Not comparable
Llama 13b: —, Mixtral 8x22B: 23.1 (#127)
| Benchmark | Llama 13b | Mixtral 8x22B |
|---|---|---|
| Cybench | — | 7.5% |
Reasoning Mixtral 8x22B leads
Llama 13b: 14.0 (#329), Mixtral 8x22B: 19.9 (#248)
| Benchmark | Llama 13b | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 728 | 1150 |
| Epoch Capabilities Index | 100.58 | 122.03 |
| DTBench | — | 55.1% |
| BIG-Bench Hard | 37.9% | — |
| ForecastBench | — | 56.3 |
| HellaSwag | 79.2% | — |
| LAMBADA | 75.2% | — |
| PIQA | 80.1% | — |
| WinoGrande | 73% | — |
Math Llama 13b leads
Llama 13b: 26.7 (#256), Mixtral 8x22B: 22.9 (#275)
| Benchmark | Llama 13b | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 838 | 1184 |
| Omni-MATH | — | 16.3% |
| MATH Level 5 | — | 24.2% |
| GSM8K | 20.6% | — |
Knowledge Not comparable
Llama 13b: —, Mixtral 8x22B: 15.1 (#293)
| Benchmark | Llama 13b | Mixtral 8x22B |
|---|---|---|
| MMLU | 47.7% | 77.8% |
| GPQA Diamond | — | 34.1% |
| MMLU-Pro | — | 46% |
| GPQA (HELM) | — | 33.4% |
| LMArena Expert | — | 1113 |
| ARC (AI2) Challenge | 52.7% | — |
| BoolQ | 78.7% | — |
| OpenBookQA | 56.4% | — |
| TriviaQA | 77.9% | — |
Multimodal Not comparable
Llama 13b: —, Mixtral 8x22B: —
| Benchmark | Llama 13b | Mixtral 8x22B |
|---|---|---|
| ScienceQA | 43.3% | — |
Multilingual Mixtral 8x22B leads
Llama 13b: 16.6 (#297), Mixtral 8x22B: 32.8 (#255)
| Benchmark | Llama 13b | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 819 | 1128 |
| LMArena Chinese | — | 1116 |
| LMArena French | — | 1166 |
| LMArena German | — | 1141 |
| LMArena Japanese | — | 1037 |
| LMArena Korean | — | 1057 |
| LMArena Russian | — | 1158 |
| LMArena Spanish | — | 1151 |
Instruction Following Mixtral 8x22B leads
Llama 13b: 36.7 (#305), Mixtral 8x22B: 57.7 (#266)
| Benchmark | Llama 13b | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 781 | 1147 |
| IFEval | — | 72.4% |
Long Context Not comparable
Llama 13b: —, Mixtral 8x22B: 34.7 (#247)
| Benchmark | Llama 13b | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | — | 1144 |
Writing & Preference Mixtral 8x22B leads
Llama 13b: 13.8 (#312), Mixtral 8x22B: 36.9 (#262)
| Benchmark | Llama 13b | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 834 | 1162 |
| LMArena Creative Writing | 794 | 1141 |
| LMArena Multi-Turn | 753 | 1130 |
| WildBench | — | 71.1% |
Frequently asked questions
Is Llama 13b better than Mixtral 8x22B?
Mixtral 8x22B is the stronger model overall, scoring 27.1 to 24.4 on the Noometry Index.
Is Llama 13b or Mixtral 8x22B better for coding?
Mixtral 8x22B scores higher on coding benchmarks: 24.2 versus 21.4 in the Noometry coding category.
How many benchmarks do Llama 13b and Mixtral 8x22B share?
10 benchmarks have published results for both models. Llama 13b has 21 scored results on Noometry and Mixtral 8x22B has 34.