Model comparison
Falcon-180B vs Mixtral 8x7B
Falcon-180B is the stronger model overall, scoring 32.2 to 27.1 on the Noometry Index.
Last verified . 14 shared benchmarks.
Technology Innovation Institute
Mistral AISummary
- They share 14 benchmarks with published results for both. Falcon-180B scores higher in 2 categories and Mixtral 8x7B in 2 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mixtral 8x7B leads 34.2 to 29.1.
Side by side
| Falcon-180B | Mixtral 8x7B | |
|---|---|---|
| Provider | Technology Innovation Institute | Mistral AI |
| Noometry Index | 32.2 | 27.1 |
| Released | 2023-09-06 | 2023-12-11 |
| Weights | Open | Open |
| Context window | — | 32K |
| Max output | — | 32K |
| Input $ / M tokens | — | $0.70 |
| Output $ / M tokens | — | $0.70 |
| Results tracked | 16 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Falcon-180B: —, Mixtral 8x7B: 32.8 (#269)
| Benchmark | Falcon-180B | Mixtral 8x7B |
|---|---|---|
| LMArena Coding | — | 1126 |
| HumanEval+ | — | 39.6% |
| MBPP+ | — | 49.7% |
Reasoning Too close to call
Falcon-180B: 19.1 (#269), Mixtral 8x7B: 18.2 (#285)
| Benchmark | Falcon-180B | Mixtral 8x7B |
|---|---|---|
| LMArena Hard Prompts | 1007 | 1115 |
| Epoch Capabilities Index | 112.13 | 118.47 |
| HellaSwag | 89% | 86.7% |
| PIQA | 84.9% | 83.6% |
| WinoGrande | 87.1% | 77.2% |
| DTBench | — | 49.6% |
| Adversarial NLI | — | 55.2% |
| ForecastBench | — | 56.3 |
| LAMBADA | 79.8% | — |
Math Not comparable
Falcon-180B: —, Mixtral 8x7B: 18.8 (#289)
| Benchmark | Falcon-180B | Mixtral 8x7B |
|---|---|---|
| GSM8K | 54.4% | 74.4% |
| Omni-MATH | — | 10.5% |
| LMArena Math | — | 1147 |
| MATH Level 5 | — | 10% |
Knowledge Not comparable
Falcon-180B: —, Mixtral 8x7B: 11.0 (#301)
| Benchmark | Falcon-180B | Mixtral 8x7B |
|---|---|---|
| ARC (AI2) Challenge | 67.8% | 87.3% |
| MMLU | 70.6% | 70.6% |
| OpenBookQA | 64.2% | 85.8% |
| GPQA Diamond | — | 30.6% |
| MMLU-Pro | — | 33.5% |
| GPQA (HELM) | — | 29.6% |
| LMArena Expert | — | 1088 |
| BoolQ | 89% | — |
| TriviaQA | — | 82.2% |
Multilingual Mixtral 8x7B leads
Falcon-180B: 25.2 (#286), Mixtral 8x7B: 29.6 (#266)
| Benchmark | Falcon-180B | Mixtral 8x7B |
|---|---|---|
| LMArena Non-English | 1000 | 1077 |
| LMArena Chinese | — | 1055 |
| LMArena French | — | 1166 |
| LMArena German | — | 1114 |
| LMArena Japanese | — | 931 |
| LMArena Korean | — | 968 |
| LMArena Russian | — | 1090 |
| LMArena Spanish | — | 1111 |
Instruction Following Falcon-180B leads
Falcon-180B: 53.4 (#286), Mixtral 8x7B: 51.0 (#297)
| Benchmark | Falcon-180B | Mixtral 8x7B |
|---|---|---|
| LMArena Instruction Following | 1047 | 1109 |
| IFEval | — | 57.5% |
Long Context Not comparable
Falcon-180B: —, Mixtral 8x7B: 33.4 (#260)
| Benchmark | Falcon-180B | Mixtral 8x7B |
|---|---|---|
| LMArena Longer Query | — | 1103 |
Writing & Preference Mixtral 8x7B leads
Falcon-180B: 29.1 (#295), Mixtral 8x7B: 34.2 (#270)
| Benchmark | Falcon-180B | Mixtral 8x7B |
|---|---|---|
| LMArena Text | 1054 | 1132 |
| LMArena Creative Writing | 1089 | 1109 |
| LMArena Multi-Turn | 1013 | 1115 |
| WildBench | — | 67.3% |
Frequently asked questions
Is Falcon-180B better than Mixtral 8x7B?
Falcon-180B is the stronger model overall, scoring 32.2 to 27.1 on the Noometry Index.
How many benchmarks do Falcon-180B and Mixtral 8x7B share?
14 benchmarks have published results for both models. Falcon-180B has 16 scored results on Noometry and Mixtral 8x7B has 38.