Sri-Vigneshwar-DJ/ibm-granite-3.0-8b-GGUF

This model was converted to GGUF format from granite-3.0-8b-instruct using llama.cpp Refer to the original model card for more details on the model.

Use with llama.cpp

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

brew install llama.cpp or !git clone https://github.com/ggerganov/llama.cpp.git

Invoke the llama.cpp server or the CLI.

or

CLI:

! /content/llama.cpp/llama-cli -m ./quantized_model/FP16.gguf/ibm-granite-3.0-8b-GGUF -n 90 --repeat_penalty 1.0 --color -i -r "User:" -f /content/llama.cpp/prompts/chat-with-bob.txt

or

llama-cli --hf-repo Sri-Vigneshwar-DJ/ibm-granite-3.0-8b-GGUF --hf-file FP16.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo Sri-Vigneshwar-DJ/ibm-granite-3.0-8b-GGUF --hf-file FP8.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 or ''!make GGML_OPENBLAS=1' along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

or

!make GGML_OPENBLAS=1

Step 3: Run inference through the main binary.

./llama-cli --hf-repo Sri-Vigneshwar-DJ/ibm-granite-3.0-8b-GGUF --hf-file FP8.gguf -p "Hi, Generate a detailed insight on 2024 Meta Campaigns"

or

./llama-server --hf-repo Sri-Vigneshwar-DJ/ibm-granite-3.0-8b-GGUF --hf-file sFP8.gguf -c 2048