metadata
base_model: bigcode/starcoder
datasets:
- bigcode/the-stack-dedup
library_name: transformers
license: bigcode-openrail-m
metrics:
- code_eval
pipeline_tag: text-generation
tags:
- code
- llama-cpp
- gguf-my-repo
inference: true
widget:
- text: 'def print_hello_world():'
example_title: Hello world
group: Python
extra_gated_prompt: >-
## Model License Agreement
Please read the BigCode [OpenRAIL-M
license](https://huggingface.co/spaces/bigcode/bigcode-model-license-agreement)
agreement before accepting it.
extra_gated_fields:
I accept the above license agreement, and will use the Model complying with the set of use restrictions and sharing requirements: checkbox
model-index:
- name: StarCoder
results:
- task:
type: text-generation
dataset:
name: HumanEval (Prompted)
type: openai_humaneval
metrics:
- type: pass@1
value: 0.408
name: pass@1
verified: false
- type: pass@1
value: 0.336
name: pass@1
verified: false
- task:
type: text-generation
dataset:
name: MBPP
type: mbpp
metrics:
- type: pass@1
value: 0.527
name: pass@1
verified: false
- task:
type: text-generation
dataset:
name: DS-1000 (Overall Completion)
type: ds1000
metrics:
- type: pass@1
value: 0.26
name: pass@1
verified: false
- task:
type: text-generation
dataset:
name: MultiPL-HumanEval (C++)
type: nuprl/MultiPL-E
metrics:
- type: pass@1
value: 0.3155
name: pass@1
verified: false
- type: pass@1
value: 0.2101
name: pass@1
verified: false
- type: pass@1
value: 0.1357
name: pass@1
verified: false
- type: pass@1
value: 0.1761
name: pass@1
verified: false
- type: pass@1
value: 0.3022
name: pass@1
verified: false
- type: pass@1
value: 0.2302
name: pass@1
verified: false
- type: pass@1
value: 0.3079
name: pass@1
verified: false
- type: pass@1
value: 0.2389
name: pass@1
verified: false
- type: pass@1
value: 0.2608
name: pass@1
verified: false
- type: pass@1
value: 0.1734
name: pass@1
verified: false
- type: pass@1
value: 0.3357
name: pass@1
verified: false
- type: pass@1
value: 0.155
name: pass@1
verified: false
- type: pass@1
value: 0.0124
name: pass@1
verified: false
- type: pass@1
value: 0.0007
name: pass@1
verified: false
- type: pass@1
value: 0.2184
name: pass@1
verified: false
- type: pass@1
value: 0.2761
name: pass@1
verified: false
- type: pass@1
value: 0.1046
name: pass@1
verified: false
- type: pass@1
value: 0.2274
name: pass@1
verified: false
- type: pass@1
value: 0.3229
name: pass@1
verified: false
osukhoroslov-hw/starcoder-Q5_K_M-GGUF
This model was converted to GGUF format from bigcode/starcoder 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 (works on Mac and Linux)
brew install llama.cpp
Invoke the llama.cpp server or the CLI.
CLI:
llama-cli --hf-repo osukhoroslov-hw/starcoder-Q5_K_M-GGUF --hf-file starcoder-q5_k_m.gguf -p "The meaning to life and the universe is"
Server:
llama-server --hf-repo osukhoroslov-hw/starcoder-Q5_K_M-GGUF --hf-file starcoder-q5_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.
./llama-cli --hf-repo osukhoroslov-hw/starcoder-Q5_K_M-GGUF --hf-file starcoder-q5_k_m.gguf -p "The meaning to life and the universe is"
or
./llama-server --hf-repo osukhoroslov-hw/starcoder-Q5_K_M-GGUF --hf-file starcoder-q5_k_m.gguf -c 2048