Instructions to use legesher/language-decoded-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use legesher/language-decoded-lora with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="legesher/language-decoded-lora")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("legesher/language-decoded-lora", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use legesher/language-decoded-lora with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "legesher/language-decoded-lora"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "legesher/language-decoded-lora",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/legesher/language-decoded-lora

SGLang

How to use legesher/language-decoded-lora with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "legesher/language-decoded-lora" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "legesher/language-decoded-lora",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "legesher/language-decoded-lora" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "legesher/language-decoded-lora",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Unsloth Studio

How to use legesher/language-decoded-lora with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for legesher/language-decoded-lora to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for legesher/language-decoded-lora to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for legesher/language-decoded-lora to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="legesher/language-decoded-lora",
    max_seq_length=2048,
)

Docker Model Runner
How to use legesher/language-decoded-lora with Docker Model Runner:
```
docker model run hf.co/legesher/language-decoded-lora
```

docs: add arxiv: tags for paper bibliography anchors

#11

by madiedgar - opened 14 days ago

base: refs/heads/main

←

from: refs/pr/11

Discussion Files changed

-0

madiedgar

Legesher org 14 days ago

Adds 8 arxiv: tags to the YAML frontmatter so the paper's load-bearing references (base model, source corpus, predecessor work, and benchmark origins) appear under the HF Papers discovery surface. Tag IDs verified against arxiv.org titles.

Tags:

arxiv:2408.10914 Aryabumi 2024 - To Code or Not to Code
arxiv:2603.11510 Salamanca 2026 - Tiny Aya (base model)
arxiv:2211.15533 Kocetkov 2022 - The Stack (source corpus)
arxiv:2510.09591 Bazaz & Beg 2025 - Multilingual Python
arxiv:1809.05053 Conneau 2018 - XNLI
arxiv:2308.16884 Bandarkar 2024 - Belebele
arxiv:2106.06937 Lin 2021 - X-CSQA
arxiv:2210.03057 Shi 2022 - MGSM (Phase 2 legacy)

SIB-200 (arxiv:2309.07445) intentionally held for a follow-up.
Tags only — no other YAML or body changes.

Summary

Adds 8 arxiv: tags to YAML frontmatter. Tags anchor the repo on the HF Papers discovery surface for each load-bearing paper in the submitted paper's bibliography.

Tag selection rationale

Tier A — load-bearing methodological foundations:

arxiv:2408.10914 Aryabumi 2024 — direct precedent (English code improves reasoning); our work extends to multilingual code
arxiv:2603.11510 Salamanca 2026 — Tiny Aya, the base model
arxiv:2211.15533 Kocetkov 2022 — The Stack, the source corpus
arxiv:2510.09591 Bazaz & Beg 2025 — Multilingual Python, predecessor to Legesher

Tier B — Phase 3 benchmark anchors:

arxiv:1809.05053 Conneau 2018 — XNLI
arxiv:2308.16884 Bandarkar 2024 — Belebele
arxiv:2106.06937 Lin 2021 — X-CSQA

Tier B legacy — Phase 2:

arxiv:2210.03057 Shi 2022 — MGSM (used in Phase 2 only; retained because Phase 2 artifacts still live in this repo)

What's NOT added

SIB-200 (arxiv:2309.07445) — verified clean, deliberately held for a follow-up batch
Background / methodology survey citations (mT5, MindMerger, Ghosh survey, etc.) — would dilute the discovery surface; remain in the paper's bibliography
Software-only references (Wenyan, Qi, Latino, Legesher itself, Edgar & Buhai 2026) — no arxiv preprints exist

Verification

All 8 IDs were fetched from arxiv.org and title-matched against the paper bibliography. No tag refers to a paper not cited in the work.

docs: add arxiv: tags for paper bibliography anchorsb1a44182

madiedgar changed pull request status to merged 14 days ago

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment