📚

Recommended Stacks

Pre-configured combinations of launcher + engine + format + quantization for your hardware.

🔴 AMD Stacks

AMD GPU users - ROCm ecosystem

16GB VRAM (RX 7900 XT)

16GB VRAM

Beginner
Stack
ollama + llama.cpp
Formats
gguf
Quantization
Q4_K_M to Q5_K_M
💡 ROCm supported. Setup more complex than NVIDIA
Power
Stack
text-generation-webui + llama.cpp
Formats
gguf
Quantization
Q5_K_M
💡 ROCm environment setup required

24GB VRAM (RX 7900 XTX)

24GB VRAM

Beginner
Stack
ollama + llama.cpp
Formats
gguf
Quantization
Q5_K_M to Q6_K
💡 70B Q4 within reach
Power
Stack
vllm + vllm
Formats
safetensors
Quantization
FP16
💡 vLLM ROCm compatible