Model comparison
Mixtral 8x7B vs Qwen1.5-14B
Qwen1.5-14B is the stronger model overall, scoring 32.1 to 26.8 on the Noometry Index.
Last verified . 17 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. Mixtral 8x7B scores higher in 1 category and Qwen1.5-14B in 7 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen1.5-14B leads 29.3 to 11.0.
Side by side
| Mixtral 8x7B | Qwen1.5-14B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 26.8 | 32.1 |
| Released | 2023-12-11 | 2024-02-04 |
| Weights | Open | Open |
| Context window | 32K | — |
| Max output | 32K | — |
| Input $ / M tokens | $0.70 | — |
| Output $ / M tokens | $0.70 | — |
| Results tracked | 38 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Mixtral 8x7B: 32.2 (#265), Qwen1.5-14B: 32.5 (#259)
| Benchmark | Mixtral 8x7B | Qwen1.5-14B |
|---|---|---|
| LMArena Coding | 1127 | 1138 |
| HumanEval+ | 39.6% | — |
| MBPP+ | 49.7% | — |
Reasoning Qwen1.5-14B leads
Mixtral 8x7B: 17.4 (#283), Qwen1.5-14B: 20.5 (#217)
| Benchmark | Mixtral 8x7B | Qwen1.5-14B |
|---|---|---|
| LMArena Hard Prompts | 1115 | 1113 |
| DTBench | 49.6% | — |
| Adversarial NLI | 55.2% | — |
| Epoch Capabilities Index | 118.38 | — |
| ForecastBench | 56.3 | — |
| HellaSwag | 86.7% | — |
| PIQA | 83.6% | — |
| WinoGrande | 77.2% | — |
Math Qwen1.5-14B leads
Mixtral 8x7B: 17.2 (#284), Qwen1.5-14B: 29.9 (#214)
| Benchmark | Mixtral 8x7B | Qwen1.5-14B |
|---|---|---|
| LMArena Math | 1147 | 1125 |
| Omni-MATH | 10.5% | — |
| MATH Level 5 | 10% | — |
| GSM8K | 74.4% | — |
Knowledge Qwen1.5-14B leads
Mixtral 8x7B: 11.0 (#296), Qwen1.5-14B: 29.3 (#230)
| Benchmark | Mixtral 8x7B | Qwen1.5-14B |
|---|---|---|
| LMArena Expert | 1088 | 1094 |
| MMLU | 70.6% | 68.6% |
| GPQA Diamond | 30.6% | — |
| MMLU-Pro | 33.5% | — |
| GPQA (HELM) | 29.6% | — |
| ARC (AI2) Challenge | 87.3% | — |
| OpenBookQA | 85.8% | — |
| TriviaQA | 82.2% | — |
Multilingual Qwen1.5-14B leads
Mixtral 8x7B: 29.8 (#261), Qwen1.5-14B: 30.8 (#257)
| Benchmark | Mixtral 8x7B | Qwen1.5-14B |
|---|---|---|
| LMArena Non-English | 1077 | 1095 |
| LMArena Chinese | 1055 | 1147 |
| LMArena French | 1165 | 1115 |
| LMArena German | 1114 | 1043 |
| LMArena Japanese | 931 | 1019 |
| LMArena Russian | 1090 | 1046 |
| LMArena Spanish | 1110 | 1085 |
| LMArena Korean | 968 | — |
Instruction Following Qwen1.5-14B leads
Mixtral 8x7B: 51.2 (#292), Qwen1.5-14B: 57.1 (#266)
| Benchmark | Mixtral 8x7B | Qwen1.5-14B |
|---|---|---|
| LMArena Instruction Following | 1110 | 1102 |
| IFEval | 57.5% | — |
Long Context Too close to call
Mixtral 8x7B: 33.4 (#255), Qwen1.5-14B: 33.7 (#252)
| Benchmark | Mixtral 8x7B | Qwen1.5-14B |
|---|---|---|
| LMArena Longer Query | 1103 | 1113 |
Writing & Preference Too close to call
Mixtral 8x7B: 34.4 (#265), Qwen1.5-14B: 33.7 (#270)
| Benchmark | Mixtral 8x7B | Qwen1.5-14B |
|---|---|---|
| LMArena Text | 1132 | 1128 |
| LMArena Creative Writing | 1109 | 1091 |
| LMArena Multi-Turn | 1115 | 1110 |
| WildBench | 67.3% | — |
Frequently asked questions
Is Mixtral 8x7B better than Qwen1.5-14B?
Qwen1.5-14B is the stronger model overall, scoring 32.1 to 26.8 on the Noometry Index.
Is Mixtral 8x7B or Qwen1.5-14B better for coding?
They score almost the same on coding (32.2 vs 32.5); test both on your own repository before choosing.
How many benchmarks do Mixtral 8x7B and Qwen1.5-14B share?
17 benchmarks have published results for both models. Mixtral 8x7B has 38 scored results on Noometry and Qwen1.5-14B has 17.