Model comparison
Claude 2 vs Phi-2
Claude 2 has enough public results to be ranked (#346); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 4 shared benchmarks.
MicrosoftSummary
- They share 4 benchmarks with published results for both.
- Phi-2 has downloadable open weights; the other is API-only.
Side by side
| Claude 2 | Phi-2 | |
|---|---|---|
| Provider | Anthropic | Microsoft |
| Noometry Index | 25.0 | — |
| Released | 2023-07-11 | 2023-12-12 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 8 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Claude 2: —, Phi-2: —
| Benchmark | Claude 2 | Phi-2 |
|---|---|---|
| HumanEval+ | 61.6% | 45.1% |
| MBPP+ | — | 54.2% |
Reasoning Not comparable
Claude 2: 21.7 (#216), Phi-2: —
| Benchmark | Claude 2 | Phi-2 |
|---|---|---|
| Epoch Capabilities Index | 120.13 | 107.94 |
| DTBench | 51.9% | — |
| Adversarial NLI | — | 42.5% |
| BIG-Bench Hard | — | 59.4% |
| HellaSwag | — | 53.6% |
| WinoGrande | — | 54.7% |
Math Not comparable
Claude 2: 9.3 (#320), Phi-2: —
| Benchmark | Claude 2 | Phi-2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | — |
| MATH Level 5 | 11.7% | — |
Knowledge Not comparable
Claude 2: 16.9 (#287), Phi-2: —
| Benchmark | Claude 2 | Phi-2 |
|---|---|---|
| MMLU | 78.5% | 58.4% |
| TriviaQA | 87.5% | 45.2% |
| GPQA Diamond | 34.7% | — |
| ARC (AI2) Challenge | — | 75.9% |
| OpenBookQA | — | 73.6% |
Frequently asked questions
Is Claude 2 better than Phi-2?
Claude 2 has enough public results to be ranked (#346); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do Claude 2 and Phi-2 share?
4 benchmarks have published results for both models. Claude 2 has 8 scored results on Noometry and Phi-2 has 11.