PsiPi/cwm-Q2_K-GGUF

Refer to the original model card for more details on the model.

Fitting on 24gb in LMStudio

• Flash attention ENABLED
• K Cache Quant type Q8_0
• V Cache Quant type Q8_0
• Layer Offload 64
• Context Length ~50k

Fitting on 24gb in LMStudio @Q4_0

• Flash attention ENABLED
• K Cache Quant type Q4_0
• V Cache Quant type Q4_0
• Layer Offload 64
• Context Length 131072

Use with llama.cpp

Install llama.cpp through brew (works on Mac and Linux)

brew install llama.cpp

Invoke the llama.cpp server or the CLI.

CLI:

llama-cli --hf-repo PsiPi/cwm-Q2_K-GGUF --hf-file cwm-q2_k.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo PsiPi/cwm-Q2_K-GGUF --hf-file cwm-q2_k.gguf -c 2048

Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.

git clone https://github.com/ggerganov/llama.cpp

Step 2: Move into the llama.cpp folder and build it with LLAMA_CURL=1 flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

Step 3: Run inference through the main binary.

./llama-cli --hf-repo PsiPi/cwm-Q2_K-GGUF --hf-file cwm-q2_k.gguf -p "The meaning to life and the universe is"

or

./llama-server --hf-repo PsiPi/cwm-Q2_K-GGUF --hf-file cwm-q2_k.gguf -c 2048