Instructions to use build-small-hackathon/agenda-parser-medium with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use build-small-hackathon/agenda-parser-medium with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="build-small-hackathon/agenda-parser-medium",
	filename="agenda-parser-medium-Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use build-small-hackathon/agenda-parser-medium with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf build-small-hackathon/agenda-parser-medium:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf build-small-hackathon/agenda-parser-medium:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf build-small-hackathon/agenda-parser-medium:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf build-small-hackathon/agenda-parser-medium: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 build-small-hackathon/agenda-parser-medium:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf build-small-hackathon/agenda-parser-medium: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 build-small-hackathon/agenda-parser-medium:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf build-small-hackathon/agenda-parser-medium:Q4_K_M

Use Docker

docker model run hf.co/build-small-hackathon/agenda-parser-medium:Q4_K_M

LM Studio
Jan

vLLM

How to use build-small-hackathon/agenda-parser-medium with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "build-small-hackathon/agenda-parser-medium"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "build-small-hackathon/agenda-parser-medium",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/build-small-hackathon/agenda-parser-medium:Q4_K_M

Ollama
How to use build-small-hackathon/agenda-parser-medium with Ollama:
```
ollama run hf.co/build-small-hackathon/agenda-parser-medium:Q4_K_M
```

Unsloth Studio

How to use build-small-hackathon/agenda-parser-medium 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 build-small-hackathon/agenda-parser-medium 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 build-small-hackathon/agenda-parser-medium to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for build-small-hackathon/agenda-parser-medium to start chatting

How to use build-small-hackathon/agenda-parser-medium with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf build-small-hackathon/agenda-parser-medium:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "build-small-hackathon/agenda-parser-medium:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use build-small-hackathon/agenda-parser-medium with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf build-small-hackathon/agenda-parser-medium: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 build-small-hackathon/agenda-parser-medium:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use build-small-hackathon/agenda-parser-medium with Docker Model Runner:
```
docker model run hf.co/build-small-hackathon/agenda-parser-medium:Q4_K_M
```

Lemonade

How to use build-small-hackathon/agenda-parser-medium with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull build-small-hackathon/agenda-parser-medium:Q4_K_M

Run and chat with the model

lemonade run user.agenda-parser-medium-Q4_K_M

List all available models

lemonade list

agenda-parser-medium

A Gemma 4 26B-A4B (MoE) fine-tune that drives the Agenda Parser agents' tool-calling loop — quantized to Q4_K_M GGUF for llama.cpp.

This is the medium member of a three-model family (26B total / ~4B active params) fine-tuned to follow a strict ReAct single-JSON-action protocol over public-meeting agenda packets and local-government legal questions. It is not a general chat assistant.


Base model	`google/gemma-4-26B-A4B-it` (26B total / ~4B active)
Method	LoRA SFT → merged → Q4_K_M GGUF
Training data	`build-small-hackathon/agenda-parser-tool-traces` (`filtered` config)
LoRA adapter	`build-small-hackathon/agenda-parser-medium-lora`
License	Gemma Terms of Use

What it does — the agent protocol

The model is trained to act as a ReAct agent that calls one tool at a time. Each step it must emit a single JSON object and nothing else:

{"thought": "<one short sentence>", "tool": "<tool name>", "args": { ... }}

It reads the tool's result, then emits the next action, until it calls final_answer. It is trained on two toolkits:

Agenda packet research — list_agenda_items, get_item_text, search_packet (semantic), find_text (exact), summarize, report, final_answer. Answers questions about an uploaded agenda packet (what an item approves, costs, dates, which items mention X, briefings).
Cornell LII legal research (scoped to local-government law) — search_regulations, resolve_cfr/resolve_usc, mcl_find/mcl_search/mcl_text/mcl_outline/mcl_lookup, etc. Answers questions on Open Meetings Act, FOIA, municipal budgeting/taxation, zoning, ethics, and the Michigan statutes governing local governments — citing CFR/USC and reading Michigan MCL text.

How it was trained

Teacher traces. Two strong teacher models — Kimi k2.6 and DeepSeek 4 pro (via OpenCode Go) — drove the real agent loop over 11 public agenda packets and a set of local-government legal questions. Tools executed live, so every observation is grounded.
Judge filtering. Each completed trace's final answer was scored for faithfulness against the text the agent actually retrieved (fast OpenCode-Go judge); only high-faithfulness traces were kept. One accepted agent step = one training example.
SFT. LoRA on the base's attention projections (q/k/v/o), 4 epochs over 974 examples (held-out packet excluded — see Evaluation), full-sequence loss (the Gemma chat template lacks {% generation %} markers for assistant-only loss), bf16 + gradient checkpointing, then merged and converted to GGUF.

hyperparameter	value
LoRA rank / α / dropout	32 / 64 / 0.05
target modules	attention + MLP `q,k,v,o,gate,up,down_proj` (auto-detected real `nn.Linear`)
epochs	4
learning rate	1e-4 (cosine, 3% warmup)
batch × grad-accum	1 × 16
max sequence length	4096
precision / GPU	bf16 / H100
final in-training token accuracy	~0.96

The full training/generation pipeline (trace capture, judge, LoRA, merge, GGUF) is reproducible from the dataset card.

Training data & provenance

Built from build-small-hackathon/agenda-parser-tool-traces: per-step {system, user, assistant} chat examples whose system message is the deployed agent's exact tool catalog + protocol. The source agenda packets are published in that dataset's source_packets/ folder; each trace row links to its source by meta.unit_id. Distilled from third-party teacher models (their terms may apply to generated text); source PDFs are public meeting records.

Sibling models

model	base	quant	this card
`agenda-parser-lite`	Gemma 4 E4B	Q8_0
`agenda-parser-medium`	Gemma 4 26B-A4B (MoE)	Q4_K_M	✅
`agenda-parser-high`	Gemma 4 26B-A4B (MoE)	Q8_0

(lite = fast/small; medium = balanced; high = best quality. medium/high share the 26B-A4B base, fine-tuned independently and shipped at different quants.)

Evaluation

One agenda packet (oakland-1570) and a held-out task seed are excluded from training and reserved for a base-vs-fine-tuned A/B benchmark (objective protocol metrics — valid-JSON-action rate, clean-final_answer rate, tool-error rate — plus an LLM-judge of answer faithfulness, absolute and pairwise). See the project repo's sft/eval.py.

Run

huggingface-cli download build-small-hackathon/agenda-parser-medium agenda-parser-medium-Q4_K_M.gguf
# --jinja loads the embedded chat/tool template
llama-server -m agenda-parser-medium-Q4_K_M.gguf --jinja

The model expects the agent's system prompt (tool catalog + protocol) and replies with one JSON action per turn.

Intended use & limitations

Intended: the in-process llama.cpp backend for the Agenda Parser agents over uploaded agenda PDFs and local-government legal lookups.
Out of scope: general-purpose chat; non-tool-calling use; legal/financial advice. Always verify answers against the cited source packet / statute.
Inherits the Gemma Terms of Use and use restrictions.