- 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
granite-4.2-30b
IBM
GermanApache 2.0
Open language model for chat and text. 3 sizes from 3.7B to 29.3B parameters.
Settings
Quality
4 bit (Q4_K_M) is the most common choice: good quality with low memory requirements.
Context length: 8k tokens
4k128k
How much text the model keeps in view at once. More context needs more memory.
Recommendation
1× H100 80 GB
The model needs about 21.2 GB of memory. Hopper, FP8 support.
- Weights
- 17.6 GB
- KV cache
- 2.1 GB
- Speed
- ~54.8 Tok./s
More suitable configurations
| Card | Quantity | Tok./s | Status |
|---|---|---|---|
| Tesla T4 16 GB | 2× | 8.8 | 2-3 weeks delivery time |
| Tesla P100 16 GB | 2× | 11.1 | 2-3 weeks delivery time |
| Tesla P40 24 GB | 1× | 11.2 | 2-3 weeks delivery time |
| Tesla V100 16 GB | 2× | 24.5 | only 1 in stock |
| Tesla V100 32 GB | 1× | 26.4 | On order, delivery time approx. 1 week |
| AMD Instinct MI50 16 GB | 2× | 18.8 | in stock |
| AMD Instinct MI50 32 GB | 1× | 20.3 | in stock |
| AMD Instinct MI60 32 GB | 1× | 18.7 | 2-3 weeks delivery time |
| AMD Instinct MI100 32 GB | 1× | 22.6 | 2-3 weeks delivery time |
| A30 24 GB | 1× | 25.6 | 2-3 weeks delivery time |
| RTX A5000 24 GB | 1× | 26.6 | 2-3 weeks delivery time |
| A40 48 GB | 1× | 25.4 | 2-3 weeks delivery time |
