• 3 % discount on prepayment
  • Insured shipping from Switzerland
  • Sustainable: refurbished instead of new
  • Warranty on all cards
← Back to overview

gemma-4-12B-it

Google
Apache 2.0

Open language model for chat and text. 5 sizes from 5.1B to 31.3B parameters, partly mixture-of-experts (only a part of the parameters is active per token).

Settings

Quality
4 bit (Q4_K_M) is the most common choice: good quality with low memory requirements.
Context length: 8k tokens
4k256k
How much text the model keeps in view at once. More context needs more memory.
Recommendation
1× H100 80 GB

The model needs about 9.2 GB of memory. Hopper, FP8 support.

Weights
7.2 GB
KV cache
0.5 GB
Speed
~137.6 Tok./s

More suitable configurations

CardQuantityTok./sStatus
Tesla T4 16 GB1×23.82-3 weeks delivery time
Tesla P100 16 GB1×30.02-3 weeks delivery time
Tesla P40 24 GB1×28.12-3 weeks delivery time
Tesla V100 16 GB1×66.3in stock
Tesla V100 32 GB1×66.3On order, delivery time approx. 1 week
AMD Instinct MI50 16 GB1×50.8in stock
AMD Instinct MI50 32 GB1×50.8in stock
AMD Instinct MI60 32 GB1×46.92-3 weeks delivery time
AMD Instinct MI100 32 GB1×56.82-3 weeks delivery time
A30 24 GB1×64.22-3 weeks delivery time
RTX A5000 24 GB1×66.82-3 weeks delivery time
A40 48 GB1×63.82-3 weeks delivery time