split-mode

Here is 1 public repository matching this topic...

Runbook Windows Vulkan : Llama 3.3 70B Q4 sur 6× AMD Polaris (split layer, ctx 25.6k) + patch TurboQuant optionnel

windows amp vulkan inference llama multi-gpu polaris rx570 rx580 kv-cache llm llama-cpp gguf llama-server openai-compatible-api turboquant amd-gcn split-mode

To associate your repository with the split-mode topic, visit your repo's landing page and select "manage topics."