What can AMD Radeon RX 9070 XT run for coding?

Build: RX 9070 XT + Ryzen 7 9700X + 32GB DDR5 (Linux/ROCm)

Memory: 16 GB VRAM + 32 GB system RAM
Runner: llama.cpp (ROCm)

Runs comfortably
97 models

Ranked by fit for coding use case + predicted speed. Click a row for VRAM breakdown.

#1CodeGemma 7B
7B
gemma
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 9.9 GBHeadroom: 6.1 GB
ollama run codegemma:7b
84
tok/s
Estimated
Weights
4.20 GB
KV cache
3.50 GB
Activations
0.22 GB
Runtime
2.00 GB
#2Qwen 3 8B
8B
qwen
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 11.0 GBHeadroom: 5.0 GB
ollama run qwen3:8b
73
tok/s
Estimated
Weights
4.80 GB
KV cache
4.00 GB
Activations
0.25 GB
Runtime
2.00 GB
#3Qwen 2.5 7B Instruct
7B
qwen
Commercial OK
Quant: Q8_0Context: 8,192VRAM: 11.0 GBHeadroom: 5.0 GB
ollama run qwen2.5:7b
48
tok/s
Estimated
Weights
8.10 GB
KV cache
0.47 GB
Activations
0.41 GB
Runtime
2.00 GB
#4StarCoder 2 3B
3B
other
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 5.6 GBHeadroom: 10.4 GB
196
tok/s
Estimated
Weights
2.00 GB
KV cache
1.50 GB
Activations
0.11 GB
Runtime
2.00 GB
#5Qwen 2.5 Coder 3B
3B
qwen
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 5.5 GBHeadroom: 10.5 GB
196
tok/s
Estimated
Weights
1.90 GB
KV cache
1.50 GB
Activations
0.10 GB
Runtime
2.00 GB
#6StarCoder 2 7B
7B
other
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 10.1 GBHeadroom: 5.9 GB
84
tok/s
Estimated
Weights
4.40 GB
KV cache
3.50 GB
Activations
0.23 GB
Runtime
2.00 GB
#7Codestral Mamba 7B
7B
mistral
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 9.9 GBHeadroom: 6.1 GB
84
tok/s
Estimated
Weights
4.20 GB
KV cache
3.50 GB
Activations
0.22 GB
Runtime
2.00 GB
#8Gervásio 8B PTPT
8B
llama
Commercial OK
Quant: Q4_K_MContext: 4,096VRAM: 8.6 GBHeadroom: 7.4 GB
73
tok/s
Estimated
Weights
4.40 GB
KV cache
2.00 GB
Activations
0.22 GB
Runtime
2.00 GB
#9OpenCoder 8B
8B
opencoder
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 10.9 GBHeadroom: 5.1 GB
73
tok/s
Estimated
Weights
4.70 GB
KV cache
4.00 GB
Activations
0.24 GB
Runtime
2.00 GB
#10Phi-3.5 Vision
4.2B
phi
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 6.7 GBHeadroom: 9.3 GB
140
tok/s
Estimated
Weights
2.50 GB
KV cache
2.10 GB
Activations
0.13 GB
Runtime
2.00 GB
#11Falcon 3 3B Instruct
3B
falcon
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 5.3 GBHeadroom: 10.7 GB
196
tok/s
Estimated
Weights
1.70 GB
KV cache
1.50 GB
Activations
0.09 GB
Runtime
2.00 GB
#12PhoGPT 4B Chat
3.7B
other
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 6.0 GBHeadroom: 10.0 GB
159
tok/s
Estimated
Weights
2.00 GB
KV cache
1.85 GB
Activations
0.11 GB
Runtime
2.00 GB

Runs with tradeoffs
98 models

Tight VRAM, partial CPU offload, or context-limited.

DeepSeek Coder V2 Lite (16B)
16B
deepseek
Commercial OK
Quant: Q4_K_MContext: 2,048VRAM: 14.0 GBHeadroom: 2.0 GB
  • Tight VRAM fit — only 2.0 GB headroom left for context growth
ollama run deepseek-coder-v2:16b
37
tok/s
Estimated
Weights
9.50 GB
KV cache
2.00 GB
Activations
0.48 GB
Runtime
2.00 GB
Qwen 3 14B
14B
qwen
Commercial OK
Quant: Q4_K_MContext: 2,048VRAM: 12.6 GBHeadroom: 3.4 GB
  • Tight VRAM fit — only 3.4 GB headroom left for context growth
ollama run qwen3:14b
42
tok/s
Estimated
Weights
8.40 GB
KV cache
1.75 GB
Activations
0.42 GB
Runtime
2.00 GB
Qwen 2.5 14B Instruct
14B
qwen
Commercial OK
Quant: Q4_K_MContext: 2,048VRAM: 13.1 GBHeadroom: 2.9 GB
  • Tight VRAM fit — only 2.9 GB headroom left for context growth
ollama run qwen2.5:14b
42
tok/s
Estimated
Weights
8.90 GB
KV cache
1.75 GB
Activations
0.45 GB
Runtime
2.00 GB
Llama 3.1 8B Instruct
8B
llama
Commercial OK
Quant: Q8_0Context: 8,192VRAM: 12.0 GBHeadroom: 4.0 GB
  • Tight VRAM fit — only 4.0 GB headroom left for context growth
ollama run llama3.1:8b
42
tok/s
Estimated
Weights
8.50 GB
KV cache
1.07 GB
Activations
0.43 GB
Runtime
2.00 GB
Muse Glimmer 30B
30B
other
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 20.6 GBHeadroom: 14.6 GB
  • Partial CPU offload: ~22% of layers run on CPU
  • CPU is the bottleneck — upgrading RAM bandwidth helps more than VRAM here
ollama run muse-glimmer
5
tok/s
Estimated
Weights
17.00 GB
KV cache
0.71 GB
Activations
0.86 GB
Runtime
2.00 GB
Codestral 22B
22B
mistral
Quant: Q4_K_MContext: 8,192VRAM: 26.7 GBHeadroom: 8.5 GB
  • Partial CPU offload: ~40% of layers run on CPU
  • CPU is the bottleneck — upgrading RAM bandwidth helps more than VRAM here
ollama run codestral:22b
5
tok/s
Estimated
Weights
13.00 GB
KV cache
11.00 GB
Activations
0.66 GB
Runtime
2.00 GB
Yi Coder 9B
9B
yi
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 12.2 GBHeadroom: 3.8 GB
  • Tight VRAM fit — only 3.8 GB headroom left for context growth
65
tok/s
Estimated
Weights
5.40 GB
KV cache
4.50 GB
Activations
0.28 GB
Runtime
2.00 GB
Qwen 2.5 Coder 32B Instruct
32B
qwen
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 24.1 GBHeadroom: 11.1 GB
  • Partial CPU offload: ~34% of layers run on CPU
  • CPU is the bottleneck — upgrading RAM bandwidth helps more than VRAM here
ollama run qwen2.5-coder:32b
4
tok/s
Estimated
Weights
19.00 GB
KV cache
2.15 GB
Activations
0.96 GB
Runtime
2.00 GB

What 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: 41 new comfortable, 116 new tradeoff

  • Qwen 3 0.6B
  • Qwen 3 1.7B
  • Gemma 3 270M
  • SmolLM2 135M 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: 63 new comfortable

  • Qwen 3 0.6B
  • Llama 3.1 8B Instruct
  • Gemma 4 12B
  • Qwen3.5 9B

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: 118 new comfortable

  • Qwen 3 0.6B
  • Llama 3.1 8B Instruct
  • Qwen 3 30B-A3B
  • Qwen 2.5 Coder 32B Instruct

Some links above are affiliate links. We may earn a commission at no extra cost to you. How we make money.

Won't run
top 5 popular models

Need more memory than you have. Shown for orientation.

DeepSeek V4 Pro (1.6T MoE)
1600B
deepseek
Commercial OK

Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.

Qwen 3.5 235B-A17B (MoE)
397B
qwen
Commercial OK

Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.

Qwen 3 235B-A22B
235B
qwen
Commercial OK

Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.

DeepSeek R1 (671B reasoning)
671B
deepseek
Commercial OK

Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.

Llama 4 Scout
109B
llama
Commercial OK

Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.

How to read these numbers

Measured here
Measured here - RunLocalAI ran this exact combo on owner hardware with public evidence.

Source-backed
Source-backed / community - a reproduced public source supports the speed, but it is not labeled as owner-measured.

Extrapolated
Extrapolated - predicted from a measured benchmark on similar-bandwidth hardware.

Estimated
Estimated - formula based on VRAM bandwidth and model architecture; not a benchmark row.

RunLocalAI Will-It-Run Framework →

Want a specific benchmark we don't have? Email Contact support and we'll prioritize it.