Instructions to use ciscoriordan/lsjpre-qwen3-8b-tiebreak with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ciscoriordan/lsjpre-qwen3-8b-tiebreak with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B")
model = PeftModel.from_pretrained(base_model, "ciscoriordan/lsjpre-qwen3-8b-tiebreak")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use ciscoriordan/lsjpre-qwen3-8b-tiebreak with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M

Use Docker

docker model run hf.co/ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M

LM Studio
Jan
Ollama
How to use ciscoriordan/lsjpre-qwen3-8b-tiebreak with Ollama:
```
ollama run hf.co/ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M
```
Unsloth Desktop

How to use ciscoriordan/lsjpre-qwen3-8b-tiebreak with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use ciscoriordan/lsjpre-qwen3-8b-tiebreak with Docker Model Runner:
```
docker model run hf.co/ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M
```

Lemonade

How to use ciscoriordan/lsjpre-qwen3-8b-tiebreak with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M

Run and chat with the model

lemonade run user.lsjpre-qwen3-8b-tiebreak-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use ciscoriordan/lsjpre-qwen3-8b-tiebreak with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use ciscoriordan/lsjpre-qwen3-8b-tiebreak with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "ciscoriordan/lsjpre-qwen3-8b-tiebreak:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

lsjpre-qwen3-8b-tiebreak

LoRA adapter that fine-tunes Qwen/Qwen3-8B as a tie-breaker for the lsjpre OV (OCR-voted) canonical-text pipeline. When the existing per-row vote across Tesseract / Qwen / Claude / GCV / Gemini / Logeion / lsj.txt / lsj.gr leaves a candidate tie or near-tie, the live pipeline POSTs a multi-letter MCQ to this model: {"prompt": "Row context: '<left>[?]<right>'\n\nCandidates for [?]:\nA) <c0>\nB) <c1>\nC) <c2>\n\nAnswer:", ...} and the model returns one capital letter naming the most plausible reading.

Versions

Tag	Description
`adapter_v1`	Original adapter trained on synthetic distractors (Latin mojibake / punct-drop / near-miss).
`main` (= v2)	Augmented training set adds 5,000 script-confusion hard negatives (e.g. Latin `Od.` vs Greek-omicron `Οd.`). Fixes the v1 regression of preferring Greek glyphs in Latin citation tokens.

To restore v1 weights: huggingface-cli download ciscoriordan/lsjpre-qwen3-8b-tiebreak --revision adapter_v1 --local-dir <dir>.

Training (v2)

Base: Qwen/Qwen3-8B, 4-bit NF4 via bitsandbytes
Targets all 7 linear layers (q/k/v/o_proj, gate/up/down_proj)
LoRA rank 16, alpha 16, dropout 0.05
1 epoch on 45,001 training examples (40,001 v1 + 5,000 hard-neg)
Best val_acc on hard-neg val: 1.0000 (at step 1000 of 2812)
Best val_acc on original val: 0.9580
Training time: ~37 min to reach best (step 1000) on a single RTX 4090; interrupted at step 1535 since val_acc had plateaued.

Evaluation

Adapter	Original val_acc (500-row cap)	Script-confusion val_acc (500 rows)
v1	0.9420	0.2360
v2	0.9580	1.0000

The hard-negative val set is in this repo as script_confusion_eval.json. Each row is a 3-way MCQ where the correct answer is a Latin-script citation token (e.g. Od., Hdt., Th.6.59) and one distractor is the same token with all uppercase Latin letters folded to their Greek visually-identical lookalikes (Οd., Ηdt., Τh.6.59). The third distractor is a longest-common-substring near-miss from the corpus.

Pipeline integration

The live pipeline gates Qwen calls behind:

LSJPRE_QWEN_TIEBREAK=1 master enable.
LSJPRE_QWEN_TIEBREAK_MARGIN near-tie margin threshold (default 1.0 = strict-tie only; recommended near-tie value is 1.2). Higher values consult Qwen on more rows; the v2 adapter is calibrated for this near-tie expansion.

Three runtime gates apply when MARGIN > 1.0:

Margin gate: skip Qwen if winner-vs-runner-up weight ratio exceeds the margin threshold.
Content-shape gate: skip Qwen if the winning candidate has no low-confidence shape signal (script-mix, Greek-Latin confusable lookalike, mid-word punctuation).
Response-confidence gate: only accept Qwen replies whose first non-whitespace alpha char is a single letter; reject hedging replies like "I think A or B", "Hmm, A maybe".

See scripts/qwen_tiebreak.py in the lsjpre repo for the full client implementation.