Model comparison
Mixtral 8x22B vs Qwen1.5-110B
Qwen1.5-110B is the stronger model overall, scoring 34.2 to 27.1 on the Noometry Index.
Last verified . 20 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Mixtral 8x22B scores higher in 0 categories and Qwen1.5-110B in 8 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen1.5-110B leads 31.2 to 15.1.
- The biggest single-benchmark swing is BigCodeBench Complete: 50.2% for Mixtral 8x22B and 44.4% for Qwen1.5-110B.
Side by side
| Mixtral 8x22B | Qwen1.5-110B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 27.1 | 34.2 |
| Released | 2024-04-17 | 2024-04-25 |
| Weights | Open | Open |
| Context window | 64K | — |
| Max output | 64K | — |
| Input $ / M tokens | $2 | — |
| Output $ / M tokens | $6 | — |
| Results tracked | 34 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5-110B leads
Mixtral 8x22B: 24.2 (#329), Qwen1.5-110B: 33.0 (#264)
| Benchmark | Mixtral 8x22B | Qwen1.5-110B |
|---|---|---|
| BigCodeBench Instruct | 40.6% | 35% |
| LMArena Coding | 1166 | 1184 |
| BigCodeBench Complete | 50.2% | 44.4% |
| WeirdML | 3.2% | — |
| HumanEval+ | 72% | — |
| MBPP+ | 64.3% | — |
Agentic & Tool Use Not comparable
Mixtral 8x22B: 23.1 (#127), Qwen1.5-110B: —
| Benchmark | Mixtral 8x22B | Qwen1.5-110B |
|---|---|---|
| Cybench | 7.5% | — |
Reasoning Qwen1.5-110B leads
Mixtral 8x22B: 19.9 (#248), Qwen1.5-110B: 22.7 (#189)
| Benchmark | Mixtral 8x22B | Qwen1.5-110B |
|---|---|---|
| LMArena Hard Prompts | 1150 | 1168 |
| ForecastBench | 56.3 | 57.7 |
| DTBench | 55.1% | — |
| Epoch Capabilities Index | 122.03 | — |
Math Qwen1.5-110B leads
Mixtral 8x22B: 22.9 (#275), Qwen1.5-110B: 33.7 (#201)
| Benchmark | Mixtral 8x22B | Qwen1.5-110B |
|---|---|---|
| LMArena Math | 1184 | 1185 |
| Omni-MATH | 16.3% | — |
| MATH Level 5 | 24.2% | — |
Knowledge Qwen1.5-110B leads
Mixtral 8x22B: 15.1 (#293), Qwen1.5-110B: 31.2 (#219)
| Benchmark | Mixtral 8x22B | Qwen1.5-110B |
|---|---|---|
| LMArena Expert | 1113 | 1144 |
| GPQA Diamond | 34.1% | — |
| MMLU-Pro | 46% | — |
| GPQA (HELM) | 33.4% | — |
| MMLU | 77.8% | — |
Multilingual Too close to call
Mixtral 8x22B: 32.8 (#255), Qwen1.5-110B: 33.6 (#250)
| Benchmark | Mixtral 8x22B | Qwen1.5-110B |
|---|---|---|
| LMArena Non-English | 1128 | 1142 |
| LMArena Chinese | 1116 | 1206 |
| LMArena French | 1166 | 1151 |
| LMArena German | 1141 | 1123 |
| LMArena Japanese | 1037 | 1074 |
| LMArena Korean | 1057 | 1044 |
| LMArena Russian | 1158 | 1118 |
| LMArena Spanish | 1151 | 1142 |
Instruction Following Qwen1.5-110B leads
Mixtral 8x22B: 57.7 (#266), Qwen1.5-110B: 60.3 (#252)
| Benchmark | Mixtral 8x22B | Qwen1.5-110B |
|---|---|---|
| LMArena Instruction Following | 1147 | 1158 |
| IFEval | 72.4% | — |
Long Context Too close to call
Mixtral 8x22B: 34.7 (#247), Qwen1.5-110B: 35.1 (#242)
| Benchmark | Mixtral 8x22B | Qwen1.5-110B |
|---|---|---|
| LMArena Longer Query | 1144 | 1157 |
Writing & Preference Qwen1.5-110B leads
Mixtral 8x22B: 36.9 (#262), Qwen1.5-110B: 38.0 (#255)
| Benchmark | Mixtral 8x22B | Qwen1.5-110B |
|---|---|---|
| LMArena Text | 1162 | 1175 |
| LMArena Creative Writing | 1141 | 1148 |
| LMArena Multi-Turn | 1130 | 1160 |
| WildBench | 71.1% | — |
Frequently asked questions
Is Mixtral 8x22B better than Qwen1.5-110B?
Qwen1.5-110B is the stronger model overall, scoring 34.2 to 27.1 on the Noometry Index.
Is Mixtral 8x22B or Qwen1.5-110B better for coding?
Qwen1.5-110B scores higher on coding benchmarks: 33.0 versus 24.2 in the Noometry coding category.
How many benchmarks do Mixtral 8x22B and Qwen1.5-110B share?
20 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and Qwen1.5-110B has 20.