Compare LLM Models
Select 2-5 models to compare side by side.
Select models to compare
0 / 5 selected
Select 2-5 models
| Attribute |
🤖
gpt2
|
|---|---|
| Company | |
| Release Date | 2019-11-05 |
| Parameters (B) | 1.5 |
| Architecture | Decoder-only Transformer (GPT) |
| Context Window | 1024 |
| Input Modalities | text |
| Open Source | Yes |
| License | mit |
| Score | 4 |
| VRAM (GB) | 3 |
| Compute | GPT-2 XL (1.5B) needs ~3 GB VRAM at FP16 for inference; base gpt2 (124M) runs in ~1.3 GB. Any modern GPU with 4 GB+ VRAM is sufficient; CPU inference is also practical. |
| Benchmark: GSM8K | 2.5 |
| Benchmark: HumanEval | 0 |
| Benchmark: MMLU | 22.4 |
| Pricing (Input) | — |
| Pricing (Output) | — |