Prompt-Guard-86M vs deberta-v3-base-prompt-injection-v2

Specs, VRAM requirements and download trends — updated 21 September 2026.

meta-llama · A
Params
280 M
Context
30d
4.4 M
llama3.1 licence · conditional
protectai · B
Params
180 M
Context
512
30d
852 K
apache-2.0 · commercial OK

Specification comparison

Differences are highlighted; identical values are muted.

Specification Prompt-Guard-86M deberta-v3-base-prompt-injection-v2
Parameters 280 M 180 M
Architecture DebertaV2ForSequenceClassification DebertaV2ForSequenceClassification
Context length 512
Licence llama3.1 apache-2.0
Commercial use Conditional Allowed
Languages 1 1
Downloads 30d 4,353,957 851,682
Downloads all time 18.1 M 7.1 M
Quantizations on the Hub SAFETENSORS SAFETENSORS
Gated Yes — accept licence No
First seen on the Hub 2024-07-21 2024-04-20

Download trend

Daily snapshots, last 56 days (28 Jul – 21 Sep)

4.6 M272 K
Prompt-Guard-86M deberta-v3-base-prompt-injection-v2

VRAM side by side

On RTX 4090 · 24 GB · 8K context unless noted

deberta-v3-base-prompt-injection-v2 · fp16 1.3 GB / 24 GB
✅ Runs comfortably
Prompt-Guard-86M · fp16 1.8 GB / 24 GB
✅ Runs comfortably

Adjacent comparisons