Model comparison
INTELLECT-1 vs phi-3-medium 14B
phi-3-medium 14B has enough public results to be ranked (#306); INTELLECT-1 does not yet, so treat this comparison as directional.
Last verified . 6 shared benchmarks.
Hugging Face
MicrosoftSummary
- They share 6 benchmarks with published results for both.
- phi-3-medium 14B has downloadable open weights; the other is API-only.
Side by side
| INTELLECT-1 | phi-3-medium 14B | |
|---|---|---|
| Provider | Hugging Face | Microsoft |
| Noometry Index | — | 29.7 |
| Released | 2024-11-29 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 7 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
INTELLECT-1: —, phi-3-medium 14B: 36.8 (#201)
| Benchmark | INTELLECT-1 | phi-3-medium 14B |
|---|---|---|
| BigCodeBench Instruct | — | 37.6% |
| BigCodeBench Complete | — | 48.7% |
Reasoning Not comparable
INTELLECT-1: —, phi-3-medium 14B: —
| Benchmark | INTELLECT-1 | phi-3-medium 14B |
|---|---|---|
| BIG-Bench Hard | 34.9% | 81.4% |
| Epoch Capabilities Index | 100.86 | 121.23 |
| HellaSwag | 71.4% | 82.4% |
| WinoGrande | 65.8% | 81.5% |
| Adversarial NLI | — | 55.8% |
Math Not comparable
INTELLECT-1: —, phi-3-medium 14B: 27.3 (#250)
| Benchmark | INTELLECT-1 | phi-3-medium 14B |
|---|---|---|
| MATH Level 5 | — | 17.6% |
| GSM8K | 38.6% | — |
Knowledge Not comparable
INTELLECT-1: —, phi-3-medium 14B: 9.1 (#306)
| Benchmark | INTELLECT-1 | phi-3-medium 14B |
|---|---|---|
| ARC (AI2) Challenge | 54.5% | 91.6% |
| MMLU | 49.9% | 78% |
| GPQA Diamond | — | 27.6% |
| OpenBookQA | — | 87.4% |
| TriviaQA | — | 73.9% |
Frequently asked questions
Is INTELLECT-1 better than phi-3-medium 14B?
phi-3-medium 14B has enough public results to be ranked (#306); INTELLECT-1 does not yet, so treat this comparison as directional.
How many benchmarks do INTELLECT-1 and phi-3-medium 14B share?
6 benchmarks have published results for both models. INTELLECT-1 has 7 scored results on Noometry and phi-3-medium 14B has 13.