DeepSeek-R1-Distill-Qwen-1.5B VRAM Calculator
Official DeepSeek-R1-Distill-Qwen-1.5B model by DeepSeek. Calculate hardware limits, context VRAM usage, and local inference requirements.
LLM (Language Model)Developer: DeepSeek
Recommended GPU: GTX 1650 4GB / RTX 3050
12 GB
32,768 tokens
Estimated Total VRAM
3.08GB
VRAM Usage Ratio26% (3.08 / 12 GB)
Memory Allocation Breakdown
Model Weights0.91 GB
KV Cache0.88 GB
CUDA Runtime1.3 GB
✓ Verified: Ready to Run
+8.9 GB headroom remaining. Inference will run smoothly without memory bottlenecks.