paraphrase-MiniLM-L6-v2 vs paraphrase-MiniLM-L3-v2

Specs, VRAM requirements and download trends — updated 6 August 2026.

sentence-transformers · A
Params
20 M
Context
512
30d
2.8 M
apache-2.0 · commercial OK
sentence-transformers · B
Params
20 M
Context
512
30d
2.6 M
apache-2.0 · commercial OK

Specification comparison

Differences are highlighted; identical values are muted.

Specification paraphrase-MiniLM-L6-v2 paraphrase-MiniLM-L3-v2
Parameters 20 M 20 M
Architecture BertModel BertModel
Context length 512 512
Licence apache-2.0 apache-2.0
Commercial use Allowed Allowed
Languages
Downloads 30d 2,755,573 2,608,179
Downloads all time 158.4 M 22.5 M
Quantizations on the Hub SAFETENSORS SAFETENSORS
Gated No No
First seen on the Hub 2022-03-02 2022-03-02

Download trend

Daily snapshots, last 10 days (28 Jul – 6 Aug)

2.8 M2.4 M
paraphrase-MiniLM-L6-v2 paraphrase-MiniLM-L3-v2

VRAM side by side

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

paraphrase-MiniLM-L6-v2 · fp16 0.6 GB / 24 GB
✅ Runs comfortably
paraphrase-MiniLM-L3-v2 · fp16 0.6 GB / 24 GB
✅ Runs comfortably

Adjacent comparisons