newsletter/granite-20b-code-instruct-Q6_K-GGUF
This model was converted to GGUF format from ibm-granite/granite-20b-code-instruct
using llama.cpp via the ggml.ai's GGUF-my-repo space.
Refer to the original model card for more details on the model.
Use with llama.cpp
Install llama.cpp through brew.
brew install ggerganov/ggerganov/llama.cpp
Invoke the llama.cpp server or the CLI.
CLI:
llama-cli --hf-repo newsletter/granite-20b-code-instruct-Q6_K-GGUF --model granite-20b-code-instruct.Q6_K.gguf -p "The meaning to life and the universe is"
Server:
llama-server --hf-repo newsletter/granite-20b-code-instruct-Q6_K-GGUF --model granite-20b-code-instruct.Q6_K.gguf -c 2048
Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well.
git clone https://github.com/ggerganov/llama.cpp && cd llama.cpp && make && ./main -m granite-20b-code-instruct.Q6_K.gguf -n 128
- Downloads last month
- 12
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for newsletter/granite-20b-code-instruct-Q6_K-GGUF
Base model
ibm-granite/granite-20b-code-base-8k
Quantized
this model
Datasets used to train newsletter/granite-20b-code-instruct-Q6_K-GGUF
Evaluation results
- pass@1 on HumanEvalSynthesis(Python)self-reported60.400
- pass@1 on HumanEvalSynthesis(Python)self-reported53.700
- pass@1 on HumanEvalSynthesis(Python)self-reported58.500
- pass@1 on HumanEvalSynthesis(Python)self-reported42.100
- pass@1 on HumanEvalSynthesis(Python)self-reported45.700
- pass@1 on HumanEvalSynthesis(Python)self-reported42.700
- pass@1 on HumanEvalSynthesis(Python)self-reported44.500
- pass@1 on HumanEvalSynthesis(Python)self-reported42.700
- pass@1 on HumanEvalSynthesis(Python)self-reported49.400
- pass@1 on HumanEvalSynthesis(Python)self-reported32.300