Hugging Peers

beecoder-220M-python-GGUF

☆ Save
by afrideva text-generation apache-2.0
436 downloads 0 likes on Hugging Face updated

Model card

BEE-spoke-data/beecoder-220M-python-GGUF

Quantized GGUF model files for beecoder-220M-python from BEE-spoke-data

Name Quant method Size
beecoder-220m-python.fp16.gguf fp16 436.50 MB
beecoder-220m-python.q2_k.gguf q2_k 94.43 MB
beecoder-220m-python.q3_k_m.gguf q3_k_m 114.65 MB
beecoder-220m-python.q4_k_m.gguf q4_k_m 137.58 MB
beecoder-220m-python.q5_k_m.gguf q5_k_m 157.91 MB
beecoder-220m-python.q6_k.gguf q6_k 179.52 MB
beecoder-220m-python.q8_0.gguf q8_0 232.28 MB

Original Model Card:

BEE-spoke-data/beecoder-220M-python

This is BEE-spoke-data/smol_llama-220M-GQA fine-tuned for code generation on:

  • filtered version of stack-smol-XL
  • deduped version of 'algebraic stack' from proof-pile-2
  • cleaned and deduped pypi (last dataset)

This model (and the base model) were both trained using ctx length 2048.

examples

Example script for inference testing: here

It has its limitations at 220M, but seems decent for single-line or docstring generation, and/or being used for speculative decoding for such purposes.

image/png

The screenshot is on CPU on a laptop.