Downloads · 30 days
87
100% of all-time downloads
coedit-small-gguf
nirajr/coedit-small-gguf
coedit-small-gguf is a text generation model from nirajr. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
GGUF quantization of jbochi/coedit-small using llama.cpp release b11149.
All-time downloads
87
Public
Repo size
82.9 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf82.9 MB · 100%
At a glance
- Task
- Text Generation
- License
- apache-2.0
- Access
- Public
- Created
- Sep 24, 2026
- Updated
- Sep 24, 2026
- SHA
- 43e73676
Try a prompt
Base models
Model card
From the Hugging Face model README
coedit-small-GGUF
GGUF quantization of jbochi/coedit-small
using llama.cpp release b11149.
Original model: https://huggingface.co/jbochi/coedit-small
(flan-t5-small fine-tuned on Grammarly's CoEdIT dataset for grammar correction; English only, 77M parameters)
Prompt format
Fix the grammar: <sentence>
Files
| Filename | Quant | Size | Notes |
|---|---|---|---|
| coedit-small-q8_0.gguf | Q8_0 | 83 MB | Near-lossless: output differed from F16 on 6 of 123 test sentences, all comma-level. |
Usage
llama-completion -m coedit-small-q8_0.gguf --temp 0 -p "Fix the grammar: i has went to store yesterday and buyed some apple"
# I have gone to the store yesterday and bought some apple.
Conversion notes
convert_hf_to_gguf.py needs spiece.model, which the source repo does not include;
the identical file from google/flan-t5-small was used. Weights are otherwise unchanged.
Credits
jbochi for the fine-tune, Grammarly for the CoEdIT dataset, and the llama.cpp project.
- Task
- Text Generation
- License
- apache-2.0
- Languages
- en
- Created
- Sep 24, 2026
- Updated
- Sep 24, 2026