With 20 GB of VRAM, the RX 7900 XT runs Qwen3 1.7B at Q4_K_M entirely on the GPU with an 8K context (about 3.2 GB needed). At Q8_0 it fits entirely.
Q4_K_M · 8K context
The model needs about 3.2 GB and fits in the 19.4 GB available, with 16.2 GB to spare.
~3.2 GB (2.8 GB–3.6 GB) needed · estimate
Fits entirely in VRAM · 16.2 GB left
EstimateData updated 24 days agoSource: Run LLM Locally engine, Hugging Face config.json + model card
What makes this estimate less reliable
| Qwen3 1.7B Q8_08K~3.7 GB · 146–240 tok/s estimate | Excellent |
|---|---|
| Qwen3 1.7B Q6_K8K~3.6 GB · 152–253 tok/s estimate | Excellent |
| Qwen3 1.7B Q5_K_M8K~3.4 GB · 161–271 tok/s estimate | Excellent |
| Qwen3 1.7B Q4_K_M8K~3.2 GB · 170–291 tok/s estimate | Excellent |
| Qwen3 1.7B IQ4_XS8K~3.1 GB · 175–303 tok/s estimate | Excellent |
| Qwen3 1.7B Q3_K_M8K~3 GB · 181–316 tok/s estimate | Excellent |
Prices and links follow the market you choose.
20 GB GDDR6 · 800 GB/s · 315 W
ExcellentRuns Qwen3 1.7B Q4_K_M entirely in memory
No current offer imported for this market: the link opens a marketplace search.
