Replies: 1 comment
-
|
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
AlphaKure
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
硬體規格:
軟體規格:
使用模型: unsloth/DeepSeek-V3-0324-GGUF Q2_K_XL量化版本 已用llama cpp合併
使用指令: python local_chat.py --model_path deepseek-ai/DeepSeek-V3-0324 --gguf_path ~/model/GGUF/DeepseekV3/ --optimize_config_path optimize/optimize_rules/rocm/DeepSeek-V3-40G.yaml --cpu_infer 94
optimize_config 參考USE AMD GPU魔改DeepSeek-V3-Chat.yaml
VRAM使用量為32GB DRAM只用了15GB
eval token只有5.115
這樣的結果應該不太正常?
Beta Was this translation helpful? Give feedback.
All reactions