v8karlo/UNCENSORED-TinyDolphin-2.8.1-1.1b-Q4_K_M-GGUF

v8karlo/TinyDolphin-2.8.1-1.1b-Q4_K_M-GGUF

UNCENSORED model. In order to make your GGUF file type go to original Tiny Dolphin repo. https://huggingface.co/cognitivecomputations/TinyDolphin-2.8-1.1b?text=My+name+is+Teven+and+I+am+a+20-year-old+college+student+from+the+University+of+Kansas.+I+have+a+passion . Copy name of the model TinyDolphin-2.8-1.1b. Go to Convert-to-GGUF repo https://huggingface.co/spaces/ggml-org/gguf-my-repo and paste model name into Hub Model ID field, choose Quantization Method and press Submit button.

This model was converted to GGUF format from cognitivecomputations/TinyDolphin-2.8.1-1.1b using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model.

Convert Safetensors to GGUF . https://huggingface.co/spaces/ggml-org/gguf-my-repo .

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 --hf-repo v8karlo/TinyDolphin-2.8.1-1.1b-Q4_K_M-GGUF --hf-file tinydolphin-2.8.1-1.1b-q4_k_m.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo v8karlo/TinyDolphin-2.8.1-1.1b-Q4_K_M-GGUF --hf-file tinydolphin-2.8.1-1.1b-q4_k_m.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.

./main --hf-repo v8karlo/TinyDolphin-2.8.1-1.1b-Q4_K_M-GGUF --hf-file tinydolphin-2.8.1-1.1b-q4_k_m.gguf -p "The meaning to life and the universe is"

or

./server --hf-repo v8karlo/TinyDolphin-2.8.1-1.1b-Q4_K_M-GGUF --hf-file tinydolphin-2.8.1-1.1b-q4_k_m.gguf -c 2048