- 3 % discount on prepayment
- Insured shipping from Switzerland
- Sustainable: refurbished instead of new
- Warranty on all cards
- 3 % discount on prepayment
- Insured shipping from Switzerland
- Sustainable: refurbished instead of new
- Warranty on all cards
← Back to overview
llama-embed-nemotron-8b
NVIDIA
GermanOpenMDW 1.1
Creates vectors for semantic search and RAG. 2 sizes from 1.1B to 7.5B parameters.
Settings
For this model we show curated VRAM values for FP16 and FP8.
Recommendation
1× Tesla V100 16 GB
The model needs about 9.5 GB of memory. Passively cooled, fan and 8-pin EPS cable required.
- Weights
- 8.0 GB
More suitable configurations
| Card | Quantity | Status |
|---|---|---|
| Tesla T4 16 GB | 1× | 2-3 weeks delivery time |
| Tesla P100 16 GB | 1× | 2-3 weeks delivery time |
| Tesla P40 24 GB | 1× | 2-3 weeks delivery time |
| Tesla V100 16 GB | 1× | in stock |
| Tesla V100 32 GB | 1× | On order, delivery time approx. 1 week |
| AMD Instinct MI50 16 GB | 1× | in stock |
| AMD Instinct MI50 32 GB | 1× | in stock |
| AMD Instinct MI60 32 GB | 1× | 2-3 weeks delivery time |
| AMD Instinct MI100 32 GB | 1× | 2-3 weeks delivery time |
| A30 24 GB | 1× | 2-3 weeks delivery time |
| RTX A5000 24 GB | 1× | 2-3 weeks delivery time |
| A40 48 GB | 1× | 2-3 weeks delivery time |
