distilgpt2 vs pythia-70m-deduped
Specs, VRAM requirements and download trends — updated 21 September 2026.
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | distilgpt2 | pythia-70m-deduped |
|---|---|---|
| Parameters | 90 M | 100 M |
| Architecture | GPT2LMHeadModel | GPTNeoXForCausalLM |
| Context length | — | 2,048 |
| Licence | apache-2.0 | apache-2.0 |
| Commercial use | Allowed | Allowed |
| Languages | 1 | 1 |
| Downloads 30d | 2,070,768 | 1,173,142 |
| Downloads all time | 256.2 M | 15.5 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2022-03-02 | 2023-02-13 |
Download trend
Daily snapshots, last 56 days (28 Jul – 21 Sep)
2.7 M660 K
distilgpt2
pythia-70m-deduped
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
pythia-70m-deduped · fp16
0.7 GB / 24 GB
distilgpt2 · fp16
0.9 GB / 24 GB