Modelpublic

reallexi/lexi-coder-v2-slm

sourceHugging Faceotherupdated 2mo agoView on Hugging Face

0likes11downloads

Model Card

reallexi/lexi-coder-v2-slm

A standalone model of 495M parameters, derived from `Qwen/Qwen2.5-0.5B-Instruct`.

The adapter has been merged into the base weights, so no PEFT adapter is needed at runtime.

Size and requirements


Parameters	495,114,112 (495M)
Weights on disk	953 MB
Trained context length	1,024 tokens
Base model	`Qwen/Qwen2.5-0.5B-Instruct`

Approximate memory to hold the weights. Add context and runtime overhead on top.

Precision	Weights
FP16 / BF16	944 MB
8-bit (Q8_0)	472 MB
4-bit (Q4KM)	260 MB

Training


Strategy	llm
Adapter	Auto LoRA
Dataset	`databricks/databricks-dolly-15k`
Samples learned	10,000 (through phase 10 of 10)
Training steps	750
Epochs	3

Usage

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("reallexi/lexi-coder-v2-slm")
tokenizer = AutoTokenizer.from_pretrained("reallexi/lexi-coder-v2-slm")

License and attribution

The effective terms are inherited from the base model and the training data, which are not necessarily the same as this project's own license. Review both before redistributing.

—Base model: `Qwen/Qwen2.5-0.5B-Instruct`

—Training data: databricks/databricks-dolly-15k

Produced by Reallexi LLC AI Model Builder from training job #1369. Core: https://llm.reallexi.io