Model comparison
Falcon-7B vs Mistral
Mistral has enough public results to be ranked (#303); Falcon-7B does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
Technology Innovation Institute
Mistral AISummary
- Falcon-7B has downloadable open weights; the other is API-only.
Side by side
| Falcon-7B | Mistral | |
|---|---|---|
| Provider | Technology Innovation Institute | Mistral AI |
| Noometry Index | — | 29.9 |
| Released | 2023-04-24 | — |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 12 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Falcon-7B: —, Mistral: 33.8 (#250)
| Benchmark | Falcon-7B | Mistral |
|---|---|---|
| LMArena Coding | — | 1162 |
Reasoning Not comparable
Falcon-7B: —, Mistral: 22.2 (#200)
| Benchmark | Falcon-7B | Mistral |
|---|---|---|
| LMArena Hard Prompts | — | 1149 |
| BIG-Bench Hard | 28.8% | — |
| Epoch Capabilities Index | 95.1 | — |
| HellaSwag | 78.1% | — |
| LAMBADA | 74.9% | — |
| PIQA | 80.3% | — |
| WinoGrande | 67.2% | — |
Math Not comparable
Falcon-7B: —, Mistral: 22.3 (#278)
| Benchmark | Falcon-7B | Mistral |
|---|---|---|
| Omni-MATH | — | 7.2% |
| LMArena Math | — | 1180 |
| GSM8K | 6.8% | — |
Knowledge Not comparable
Falcon-7B: —, Mistral: 16.6 (#288)
| Benchmark | Falcon-7B | Mistral |
|---|---|---|
| MMLU-Pro | — | 27.7% |
| GPQA (HELM) | — | 30.3% |
| LMArena Expert | — | 1125 |
| ARC (AI2) Challenge | 47.9% | — |
| BoolQ | 75.3% | — |
| MMLU | 35% | — |
| OpenBookQA | 51.6% | — |
| TriviaQA | 64.6% | — |
Multilingual Not comparable
Falcon-7B: —, Mistral: 32.8 (#254)
| Benchmark | Falcon-7B | Mistral |
|---|---|---|
| LMArena Non-English | — | 1129 |
| LMArena Chinese | — | 1109 |
| LMArena French | — | 1180 |
| LMArena German | — | 1155 |
| LMArena Japanese | — | 1013 |
| LMArena Korean | — | 1032 |
| LMArena Russian | — | 1168 |
| LMArena Spanish | — | 1143 |
Instruction Following Not comparable
Falcon-7B: —, Mistral: 52.6 (#288)
| Benchmark | Falcon-7B | Mistral |
|---|---|---|
| IFEval | — | 56.8% |
| LMArena Instruction Following | — | 1152 |
Long Context Not comparable
Falcon-7B: —, Mistral: 35.0 (#245)
| Benchmark | Falcon-7B | Mistral |
|---|---|---|
| LMArena Longer Query | — | 1153 |
Writing & Preference Not comparable
Falcon-7B: —, Mistral: 37.0 (#260)
| Benchmark | Falcon-7B | Mistral |
|---|---|---|
| LMArena Text | — | 1165 |
| LMArena Creative Writing | — | 1158 |
| WildBench | — | 66% |
| LMArena Multi-Turn | — | 1147 |
Frequently asked questions
Is Falcon-7B better than Mistral?
Mistral has enough public results to be ranked (#303); Falcon-7B does not yet, so treat this comparison as directional.
How many benchmarks do Falcon-7B and Mistral share?
0 benchmarks have published results for both models. Falcon-7B has 12 scored results on Noometry and Mistral has 22.