SLM PDF Chat
Securely parses complex PDF documents. Assembles layouts, reads tables, and lets you chat with local legal contracts, research articles, or receipts.
01 / Overview & Capabilities
Securely parses complex PDF documents. Assembles layouts, reads tables, and lets you chat with local legal contracts, research articles, or receipts.
Key Features
- Locally extracts layout text and multi-column paragraphs
- Parses database tables inside PDFs directly to list-of-dicts
- Built-in RAG chunk generator for offline querying
- Supports scanned image PDFs via local OCR integration
02 / Standalone OS Harness Installers, CLI & Python SDK
In SLMAgents 2.0, SLM PDF Chat executes inside the 10-Plane Pure-CPU Agent Harness (slm_harness). Download a self-extracting Individual Agent + Harness OS Installer for macOS, Windows, or Linux (bundling slm_pdf + slm_harness + the ./slm CLI), install the full 26-agent desktop suite, or install via Python pip:
1. Individual SLM PDF Chat + 10-Plane Harness Installers (All OSes)
slm_pdf + slm_harness + ./slm CLIslm_pdf + slm_harnessslm_pdf + slm_harness + ./slm CLI2. Full 26-Agent Studio & Desktop OS Installers
3. 10-Plane Harness CLI (./slm) & Python pip Installation
# Option A: Install via PyPI or editable source with the 10-Plane Harness $ pip install slm-pdf $ pip install -e ./slm_harness -e ./slm_pdf # Option B: Run directly through the Unified 10-Plane Pure-CPU Harness CLI (./slm) $ ./slm ask "What are the indemnification clauses in @contract.pdf? Cite exact page numbers." $ ./slm chat # Interactive Rich TUI with 18 slash commands (/trace, /models, /arena, /undo) $ ./slm serve # 100% Offline Web AI Studio + OpenAI /v1 API at http://127.0.0.1:7860
03 / 10-Plane Pure-CPU Harness Integration (slm_harness)
Rather than executing as an isolated prompt wrapper, SLM PDF Chat is natively orchestrated by the SLMAgents 2.0 10-Plane Pure-CPU Harness (via slm_harness.runtime.legacy_bridge and CascadeRouter). Every call benefits from hybrid llama.cpp GGUF (Q4_K_M) + Quantized INT4/INT8 CPU execution, watermarked LRU memory governance, grammar-constrained decoding, and deterministic verification at $0 API cost:
| Harness Plane | Subsystem | How It Powers SLM PDF Chat on Pure CPU |
|---|---|---|
| Plane 6 | CascadeRouter (<35ms) |
Stage 1 (.pdf Attachment Detection) & Stage 2 (Page-Cited Q&A Intent) |
| Plane 2 & 5 | MemoryArena & Hybrid CPU Runtime |
Specialist Role: embeddings (mxbai-embed-large) + reranker (bge-reranker-base) + primary (Qwen3-4B GGUF Q4_K_M). Managed inside Watermark LRU MemoryArena (lite 2.6 GB / balanced 4.7 GB / power 8.2 GB) with zero-copy session sharing and automatic OOM tier step-down (8B -> 4B -> 3B -> 0.8B). |
| Plane 7 & 8 | Deterministic Verifier & Governance | Plane 7 Page-Number & Span Citation Grounding Verifier + Plane 3 Artifact Handle Indexing |
| Plane 1, 3 & 4 | Kernel, Artifact Scratchpad & 4-Tier Memory | SQLite WAL CheckpointStore (/fork, /rewind, /undo), content-addressed artifact://<sha12> handles to prevent prompt bloat, and hybrid SQLite FTS5 BM25 + HNSW vector memory. |
04 / Source Checkout (Git Sparse-Checkout)
Clone this agent alongside the 10-Plane Pure-CPU Harness (slm_harness) from the monorepo using Git sparse-checkout — no need to download all 26 agents:
Option 1 — Sparse Checkout with 10-Plane Harness (Recommended)
# 1. Create and enter a new directory $ mkdir slm_pdf_workspace && cd slm_pdf_workspace # 2. Initialise empty git repo and add remote $ git init $ git remote add origin https://github.com/t00114218-stack/SLMAgents.git # 3. Enable sparse-checkout for both slm_harness and slm_pdf $ git sparse-checkout init --cone $ git sparse-checkout set slm_harness slm_pdf # 4. Pull the 10-Plane Harness and agent source $ git pull origin main
Option 2 — Full Repository Clone
$ git clone https://github.com/t00114218-stack/SLMAgents.git $ cd SLMAgents $ ./slm doctor
Note: After checkout, install the harness and agent in editable mode with pip install -e ./slm_harness -e ./slm_pdf or download the pre-packaged OS Harness Installer above.
05 / Configuration & API Reference
Constructor Parameters
Instantiate SLMPDFChat with performance options:
| Parameter | Type / Default | Description |
|---|---|---|
| model_path | str | None | Explicit path to GGUF (Q4_K_M) or quantized ONNX model weights. If omitted, managed automatically by the 10-Plane Harness MemoryArena. |
| cache_dir | str | None | Directory to store model weights offline. Defaults to ~/.cache/slm-pdf/. Also settable via SLM_PDF_CHAT_CACHE_DIR. |
| n_threads | int | 4 | CPU thread count for 10-Plane Harness inference (auto-tuned to physical P-cores). Optimize for CPU core count. Also settable via SLM_PDF_CHAT_N_THREADS. |
Methods
| Method Signature | Return Type | Description |
|---|---|---|
load(pdf_path) | None | Saves document configurations locally and performs text extraction mappings. |
ask(question, system_prompt=None, user_input=None) | str | Executes vector search on the document text chunks to synthesize local responses. |
Method Parameters (Execution Customization)
All main execution methods accept optional system routing parameters:
| Parameter | Type / Default | Description |
|---|---|---|
| system_prompt | str | None | Optional custom system prompt instruction to override the default system template response parameters. |
| user_input | str | None | Optional additional user-supplied target text variables or contextual keys. |
Quick Start
from slm_pdf import SLMPDFChat
pdf = SLMPDFChat()
pdf.load("invoice.pdf")
ans = pdf.ask(
"What is the total due amount?",
system_prompt="Answer format: $XX.XX",
user_input="Extract tax detail explicitly"
)
print(ans)
Environment Variables
Configure agent parameters globally using environment values:
| Environment Variable | Default | Purpose |
|---|---|---|
| SLM_PDF_CHAT_N_THREADS | 4 | Sets CPU inference execution threads. |
| SLM_PDF_CHAT_CACHE_DIR | ~/.cache/slm-pdf/ | Default directory to store local GGUF (Q4_K_M) and quantized INT4/INT8 model weights (shared with ~/.slm_harness/models). |
CPU Performance Tuning
To run the SLMPDFChat engine efficiently on CPU under 1.5 GB memory footprint:
- Match Threads to Core Count: Set
n_threadsorSLM_PDF_CHAT_N_THREADSto match the physical CPU core count. - Sequential Processing: Avoid concurrent processing when batch files are large.
- Garbage Collection: Clear variables and run
gc.collect()to release model RAM blocks after execution.
Verified Input & Output Logs
Diagnostic execution console response running locally on CPU:
→ INPUT (Ask before load):
"What is total revenue?"
← OUTPUT:
"No PDF document loaded. Please call .load(pdf_path) first."