Model comparison
Falcon-40B vs Phi-2
Neither Falcon-40B nor Phi-2 has enough public benchmark results to be ranked yet; the rows below show what has been published.
Last verified . 8 shared benchmarks.
Technology Innovation Institute
MicrosoftSummary
- They share 8 benchmarks with published results for both.
Side by side
| Falcon-40B | Phi-2 | |
|---|---|---|
| Provider | Technology Innovation Institute | Microsoft |
| Noometry Index | — | — |
| Released | 2023-03-15 | 2023-12-12 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 12 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Falcon-40B: —, Phi-2: —
| Benchmark | Falcon-40B | Phi-2 |
|---|---|---|
| HumanEval+ | — | 45.1% |
| MBPP+ | — | 54.2% |
Reasoning Not comparable
Falcon-40B: —, Phi-2: —
| Benchmark | Falcon-40B | Phi-2 |
|---|---|---|
| BIG-Bench Hard | 37.1% | 59.4% |
| Epoch Capabilities Index | 104.45 | 107.94 |
| HellaSwag | 85.3% | 53.6% |
| WinoGrande | 76.9% | 54.7% |
| Adversarial NLI | — | 42.5% |
| LAMBADA | 77.3% | — |
| PIQA | 83% | — |
Math Not comparable
Falcon-40B: —, Phi-2: —
| Benchmark | Falcon-40B | Phi-2 |
|---|---|---|
| GSM8K | 33.8% | — |
Knowledge Not comparable
Falcon-40B: —, Phi-2: —
| Benchmark | Falcon-40B | Phi-2 |
|---|---|---|
| ARC (AI2) Challenge | 61.9% | 75.9% |
| MMLU | 56.9% | 58.4% |
| OpenBookQA | 56.6% | 73.6% |
| TriviaQA | 79.9% | 45.2% |
| BoolQ | 83.1% | — |
Frequently asked questions
Is Falcon-40B better than Phi-2?
Neither Falcon-40B nor Phi-2 has enough public benchmark results to be ranked yet; the rows below show what has been published.
How many benchmarks do Falcon-40B and Phi-2 share?
8 benchmarks have published results for both models. Falcon-40B has 12 scored results on Noometry and Phi-2 has 11.