Porovnať
Llama Guard 3 8B vs Llama Guard 4 12B
Postavte dva až štyri modely vedľa seba podľa polí, ktoré rozhodujú o nasadení: kontext, cena, región, licencia, jazyky, hosťovanie a čo je overené.
| Llama Guard 3 8B | Llama Guard 4 12B | |
|---|---|---|
| Poskytovateľ | Meta | Meta |
| Parametre | 8B | 12B |
| Kontextové okno | Neoverené: poskytovateľ tento údaj nedokumentoval. | Neoverené: poskytovateľ tento údaj nedokumentoval. |
| Licencia | llama3.1 | other |
| Modalitá | text | text, vision |
| Dátum vydania | 22. 7. 2024 | 23. 4. 2025 |
| Jazyky | Silná: en Dostatočná: — | Silná: en Dostatočná: — |
| Otvorené váhy | áno | áno |
| Úlohy | Citlivé údaje, Zákaznícka podpora | Citlivé údaje, Videnie, Zákaznícka podpora |
Ak sa dva modely líšia, táto tabuľka zobrazí rozdiel a skončí. LLM EU nezverejňuje odznak víťaza bez zdokumentovaného skóre pre danú úlohu a chýbajúci údaj zostáva chýbajúci, namiesto toho, aby sa stal nulou.
Kedy ho použiť
Llama Guard 3 8B
Silné stránky
- open weights are published on Hugging Face and can be downloaded without a vendor agreement
- at 8B parameters it fits on a single accelerator
- the model page lists en
- instruction-tuned checkpoint with a chat template in the repository
Obmedzenia
- catalogue only — LLM EU does not run this model, and nothing on this card implies an LLM EU endpoint
- no evaluation has been run by LLM EU
- the published configuration records no context window, so the card carries none
- at 8B parameters, the weights fit on a single modest accelerator
- a classifier, not a chat model: it labels prompts and responses against a safety taxonomy instead of answering them
Úlohy
Llama Guard 4 12B
Silné stránky
- open weights are published on Hugging Face and can be downloaded without a vendor agreement
- at 12B parameters it fits on a single accelerator
- the model page lists en
- instruction-tuned checkpoint with a chat template in the repository
Obmedzenia
- catalogue only — LLM EU does not run this model, and nothing on this card implies an LLM EU endpoint
- no evaluation has been run by LLM EU
- the llama4 (other) licence is a custom agreement whose commercial terms were not established from its published text, so commercial serving is not permitted until a human clears it; the router must not route to it
- the published configuration records no context window, so the card carries none
- at 12B parameters, self-hosting needs a single accelerator with enough memory for the weights
- vision checkpoints are heavier to serve than their parameter count alone suggests, because the image encoder and pre-processing run alongside the language model
- a classifier, not a chat model: it labels prompts and responses against a safety taxonomy instead of answering them
Úlohy
Úlohy: Citlivé údaje, Zákaznícka podpora