Go to file

mirivlad f63125e945 many fixes		2026-05-23 18:18:58 +08:00
config	Integrate memory policy and reflection runtime	2026-05-21 22:40:11 +08:00
docs	many fixes	2026-05-23 18:18:58 +08:00
duck_core	many fixes	2026-05-23 18:18:58 +08:00
prompts/roles	many fixes	2026-05-23 18:18:58 +08:00
scripts	many fixes	2026-05-23 18:18:58 +08:00
skills/analyze_project	Replace repository with DuckLM runtime	2026-05-20 01:00:28 +08:00
tests/smoke	many fixes	2026-05-23 18:18:58 +08:00
.env.example	Stabilize live chat tool loop	2026-05-22 21:13:58 +08:00
.gitignore	Replace repository with DuckLM runtime	2026-05-20 01:00:28 +08:00
CURRENT_STATE.md	many fixes	2026-05-23 18:18:58 +08:00
Ducklm.md	Replace repository with DuckLM runtime	2026-05-20 01:00:28 +08:00
Makefile	Add DuckLM service scripts and utility model benchmark	2026-05-22 07:27:59 +08:00
README.md	many fixes	2026-05-23 18:18:58 +08:00
docker-compose.memory.yml	Replace repository with DuckLM runtime	2026-05-20 01:00:28 +08:00
favicon.ico	fixes	2026-05-17 23:09:56 +08:00
pyproject.toml	Integrate memory policy and reflection runtime	2026-05-21 22:40:11 +08:00

README.md

DuckLM

DuckLM is a local agent runtime over local language models. It exposes a WebChat and HTTP API, calls llama-server through an OpenAI-compatible API, and persists tasks/events in SQLite.

Quick Start

python3 -m venv .venv
. .venv/bin/activate
python -m pip install -e ".[dev]"
cp .env.example .env
bash scripts/duck.sh start

Open http://127.0.0.1:8000/.

duck.sh starts and stops the local Qdrant vector memory service together with llama-server and the DuckLM API. Use status --probe for live backend checks.

Useful commands:

bash scripts/duck.sh status
bash scripts/duck.sh status --probe
bash scripts/duck.sh logs --follow
bash scripts/duck.sh restart
bash scripts/duck.sh stop

MTP/speculative variant:

bash scripts/duck-mtp.sh start
bash scripts/duck-mtp.sh status
bash scripts/duck-mtp.sh logs --follow
bash scripts/duck-mtp.sh stop