Qwen2.5 14B Instruct at Q4_K_M needs about 11.2 GB, more than the 11 GB of the RTX 2080 Ti: part of the model runs from system RAM, which slows generation. At Q3_K_M it fits entirely.
Q4_K_M · 8K context
The model needs about 11.2 GB but only 10.7 GB of VRAM is usable: 0.6 GB must be loaded into system RAM. It runs, 95% on the GPU, but more slowly.
~11.2 GB (10.7 GB–11.8 GB) needed · estimate
0.6 GB must be offloaded to system RAM
EstimateData updated 24 days agoSource: Run LLM Locally engine, Hugging Face config.json + model card
What makes this estimate less reliable
| Qwen2.5 14B Instruct Q8_08K~17.8 GB · 2.5–5.5 tok/s estimate | CPU/GPU offload |
|---|---|
| Qwen2.5 14B Instruct Q6_K8K~14.3 GB · 5–10 tok/s estimate | CPU/GPU offload |
| Qwen2.5 14B Instruct Q5_K_M8K~12.7 GB · 8.5–16 tok/s estimate | CPU/GPU offload |
| Qwen2.5 14B Instruct Q4_K_M8K~11.2 GB · 19–34 tok/s estimate | CPU/GPU offload |
| Qwen2.5 14B Instruct IQ4_XS8K~10.4 GB · 46–64 tok/s estimate | Compatible with limits |
| Qwen2.5 14B Instruct Q3_K_M8K~9.6 GB · 50–70 tok/s estimate | Compatible |
Prices and links follow the market you choose.
NVIDIA
ExcellentRuns Qwen2.5 14B Instruct Q4_K_M entirely in memory
AMD
ExcellentRuns Qwen2.5 14B Instruct Q4_K_M entirely in memory
Apple Silicon
CompatibleRuns Qwen2.5 14B Instruct Q4_K_M entirely in memory
Unified memory PC
ExcellentRuns Qwen2.5 14B Instruct Q4_K_M entirely in memory
No current offer imported for this market: the link opens a marketplace search.
