← Back to Models
DeepSeek V3
A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.
deepseek-v3
671B parameters
Base
Pull this model
Use the following command with the HoML CLI:
homl pull deepseek-v3:baseResource Requirements
| Quantization | Disk Space | GPU Memory |
|---|---|---|
| BF16 | 1.4 TB | 1.4 TB |