What can AMD Radeon RX 9070 XT run?
Build: AMD Radeon RX 9070 XT + — + 32 GB RAM (windows)
Runs comfortably154 models
Full-VRAM resident, with room for context. No compromises.
Quant: Q4_K_MContext: 8,192VRAM: 1.6 GBHeadroom: 14.4 GB800tok/sEstimated
Quant: Q8_0Context: 8,192VRAM: 11.0 GBHeadroom: 5.0 GBollama run llama3.1:8b34tok/sEstimated
ollama run llama3.1:8bQuant: Q4_K_MContext: 8,192VRAM: 10.0 GBHeadroom: 6.0 GBollama run qwen3:8b60tok/sEstimated
ollama run qwen3:8bQuant: Q8_0Context: 8,192VRAM: 6.1 GBHeadroom: 9.9 GBollama run llama3.2:3b91tok/sEstimated
ollama run llama3.2:3bQuant: Q4_K_MContext: 8,192VRAM: 2.8 GBHeadroom: 13.2 GB282tok/sEstimated
Quant: Q4_K_MContext: 2,048VRAM: 11.6 GBHeadroom: 4.4 GBollama run qwen3:14b34tok/sEstimated
ollama run qwen3:14bQuant: Q8_0Context: 8,192VRAM: 10.0 GBHeadroom: 6.0 GBollama run qwen2.5:7b39tok/sEstimated
ollama run qwen2.5:7bQuant: Q4_K_MContext: 8,192VRAM: 1.3 GBHeadroom: 14.7 GB1778tok/sEstimated
Quant: Q4_K_MContext: 8,192VRAM: 9.4 GBHeadroom: 6.6 GBollama run deepseek-r1:7b69tok/sEstimated
ollama run deepseek-r1:7bQuant: Q4_K_MContext: 2,048VRAM: 11.6 GBHeadroom: 4.4 GBollama run phi4:14b34tok/sEstimated
ollama run phi4:14bQuant: Q4_K_MContext: 8,192VRAM: 10.2 GBHeadroom: 5.8 GBollama run hermes3:8b60tok/sEstimated
ollama run hermes3:8bQuant: Q4_K_MContext: 8,192VRAM: 10.3 GBHeadroom: 5.7 GBollama run dolphin3:8b60tok/sEstimated
ollama run dolphin3:8bRuns with tradeoffs83 models
Tight VRAM, partial CPU offload, or context-limited.
Quant: Q4_K_MContext: 8,192VRAM: 34.9 GBHeadroom: 0.3 GB- • Partial CPU offload: ~54% of layers run on CPU
ollama run qwen3:30b16tok/sEstimated
- • Partial CPU offload: ~54% of layers run on CPU
ollama run qwen3:30bQuant: Q4_K_MContext: 8,192VRAM: 23.1 GBHeadroom: 12.1 GB- • Partial CPU offload: ~31% of layers run on CPU
ollama run qwen2.5-coder:32b15tok/sEstimated
- • Partial CPU offload: ~31% of layers run on CPU
ollama run qwen2.5-coder:32bQuant: Q4_K_MContext: 8,192VRAM: 32.4 GBHeadroom: 2.8 GB- • Partial CPU offload: ~51% of layers run on CPU
ollama run qwen3.6:27b18tok/sEstimated
- • Partial CPU offload: ~51% of layers run on CPU
ollama run qwen3.6:27bQuant: Q4_K_MContext: 2,048VRAM: 25.0 GBHeadroom: 10.2 GB- • Partial CPU offload: ~36% of layers run on CPU
ollama run qwen3:32b15tok/sEstimated
- • Partial CPU offload: ~36% of layers run on CPU
ollama run qwen3:32bQuant: Q4_K_MContext: 2,048VRAM: 23.8 GBHeadroom: 11.4 GB- • Partial CPU offload: ~33% of layers run on CPU
ollama run gemma4:31b15tok/sEstimated
- • Partial CPU offload: ~33% of layers run on CPU
ollama run gemma4:31bQuant: Q4_K_MContext: 8,192VRAM: 15.0 GBHeadroom: 1.0 GB- • Tight VRAM fit — only 1.0 GB headroom left for context growth
ollama run gemma4:12b40tok/sEstimated
- • Tight VRAM fit — only 1.0 GB headroom left for context growth
ollama run gemma4:12bQuant: Q4_K_MContext: 8,192VRAM: 12.4 GBHeadroom: 3.6 GB- • Tight VRAM fit — only 3.6 GB headroom left for context growth
ollama run qwen3.5:9b53tok/sEstimated
- • Tight VRAM fit — only 3.6 GB headroom left for context growth
ollama run qwen3.5:9bQuant: Q4_K_MContext: 2,048VRAM: 24.7 GBHeadroom: 10.5 GB- • Partial CPU offload: ~35% of layers run on CPU
ollama run qwen3-coder:30b16tok/sEstimated
- • Partial CPU offload: ~35% of layers run on CPU
ollama run qwen3-coder:30bWhat if you upgraded?
Hypothetical scenarios. We re-ran the compatibility engine for each.
+32 GB system RAM
~$80–150
Doubles your CPU-offload working set. Helps when models don't quite fit in VRAM.
Unlocks: 102 new tradeoff
- • Qwen 3 30B-A3B
- • Qwen 2.5 Coder 32B Instruct
- • Qwen3.6 27B
- • Llama 3.3 70B Instruct
Upgrade to AMD Radeon RX 7900 XT
~$729
20 GB VRAM (vs your 16 GB) plus a bandwidth jump from ~644 GB/s to ~800 GB/s.
Unlocks: 15 new comfortable
- • Gemma 4 12B
- • Qwen3.5 9B
- • Llama 3.2 11B Vision Instruct
- • Gemma 3 12B
Add a second AMD Radeon RX 9070 XT
~$649
Tensor parallelism splits the model across both cards, effectively doubling VRAM. Bandwidth doesn't double — runs ~1.5× the single-card speed in practice.
Unlocks: 71 new comfortable
- • Qwen 3 30B-A3B
- • Qwen 2.5 Coder 32B Instruct
- • Qwen3.6 27B
- • Qwen 3 32B
Some links above are affiliate links. We may earn a commission at no extra cost to you. How we make money.
Won't runtop 5 popular models
Need more memory than you have. Shown for orientation.
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
—
Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.
How to read these numbers
Want a specific benchmark we don't have? Email Contact support and we'll prioritize it.