Package Details: paper2slides-git r25.941695a-2

Git Clone URL: https://aur.archlinux.org/paper2slides-git.git (read-only, click to copy)
Package Base: paper2slides-git
Description: From paper to presentation in one click (CLI)
Upstream URL: https://github.com/HKUDS/Paper2Slides
Licenses: MIT
Conflicts: paper2slides
Provides: paper2slides
Submitter: guglovich
Maintainer: guglovich
Last Packager: guglovich
Votes: 0
Popularity: 0.000000
First Submitted: 2026-04-22 16:49 (UTC)
Last Updated: 2026-04-22 18:17 (UTC)

Pinned Comments

guglovich commented on 2026-04-22 18:19 (UTC)

paper2slides-git — Quick Start
1) Installation
yay -S paper2slides-git

The package installs the CLI and source code to /usr/lib/paper2slides, but some upstream Python dependencies are not yet packaged in the official Arch repositories (e.g., lightrag-hku, mineru[core]).


2) Recommended setup via venv (to install all dependencies)
python -m venv ~/.local/share/paper2slides/venv
source ~/.local/share/paper2slides/venv/bin/activate
pip install --upgrade pip
pip install -r /usr/lib/paper2slides/requirements.txt

3) Setting up API keys

There are two options.

Option A (persistent file):

Edit the file:

sudo nano /usr/lib/paper2slides/.env

Minimal example:

RAG_LLM_API_KEY=sk-...
RAG_LLM_BASE_URL=https://api.openai.com/v1
LLM_MODEL=gpt-4o-mini

IMAGE_GEN_PROVIDER=openrouter
IMAGE_GEN_API_KEY=sk-or-...
IMAGE_GEN_BASE_URL=https://openrouter.ai/api/v1
IMAGE_GEN_MODEL=google/gemini-3-pro-image-preview

Option B (via console, without editing files):

export RAG_LLM_API_KEY='sk-...'
export RAG_LLM_BASE_URL='https://api.openai.com/v1'
export LLM_MODEL='gpt-4o-mini'

export IMAGE_GEN_PROVIDER='openrouter'
export IMAGE_GEN_API_KEY='sk-or-...'
export IMAGE_GEN_BASE_URL='https://openrouter.ai/api/v1'
export IMAGE_GEN_MODEL='google/gemini-3-pro-image-preview'

To avoid entering them every time, add these export commands to your ~/.bashrc.


4) First run
source ~/.local/share/paper2slides/venv/bin/activate
paper2slides --input ./paper.pdf --output slides --length medium --style academic

Parallel mode:

paper2slides --input ./paper.pdf --output slides --parallel 2

5) Important
  • Valid LLM/vision provider keys are required; without them, the pipeline will stop.
  • The first run may take a while due to RAG preparation and dependency installation.
  • To force a restart of specific stages, use --from-stage rag|summary|plan|generate.

Latest Comments

guglovich commented on 2026-04-22 18:19 (UTC)

paper2slides-git — Quick Start
1) Installation
yay -S paper2slides-git

The package installs the CLI and source code to /usr/lib/paper2slides, but some upstream Python dependencies are not yet packaged in the official Arch repositories (e.g., lightrag-hku, mineru[core]).


2) Recommended setup via venv (to install all dependencies)
python -m venv ~/.local/share/paper2slides/venv
source ~/.local/share/paper2slides/venv/bin/activate
pip install --upgrade pip
pip install -r /usr/lib/paper2slides/requirements.txt

3) Setting up API keys

There are two options.

Option A (persistent file):

Edit the file:

sudo nano /usr/lib/paper2slides/.env

Minimal example:

RAG_LLM_API_KEY=sk-...
RAG_LLM_BASE_URL=https://api.openai.com/v1
LLM_MODEL=gpt-4o-mini

IMAGE_GEN_PROVIDER=openrouter
IMAGE_GEN_API_KEY=sk-or-...
IMAGE_GEN_BASE_URL=https://openrouter.ai/api/v1
IMAGE_GEN_MODEL=google/gemini-3-pro-image-preview

Option B (via console, without editing files):

export RAG_LLM_API_KEY='sk-...'
export RAG_LLM_BASE_URL='https://api.openai.com/v1'
export LLM_MODEL='gpt-4o-mini'

export IMAGE_GEN_PROVIDER='openrouter'
export IMAGE_GEN_API_KEY='sk-or-...'
export IMAGE_GEN_BASE_URL='https://openrouter.ai/api/v1'
export IMAGE_GEN_MODEL='google/gemini-3-pro-image-preview'

To avoid entering them every time, add these export commands to your ~/.bashrc.


4) First run
source ~/.local/share/paper2slides/venv/bin/activate
paper2slides --input ./paper.pdf --output slides --length medium --style academic

Parallel mode:

paper2slides --input ./paper.pdf --output slides --parallel 2

5) Important
  • Valid LLM/vision provider keys are required; without them, the pipeline will stop.
  • The first run may take a while due to RAG preparation and dependency installation.
  • To force a restart of specific stages, use --from-stage rag|summary|plan|generate.