SLM Summarizer
Local CPU text distillation engine that compresses long-form documents, articles, and financial notes into structured bullet highlights.
01 / Overview & Capabilities
Local CPU text distillation engine that compresses long-form documents, articles, and financial notes into structured bullet highlights.
Key Features
- Sub-billion parameter local distillation model
- High compression ratio with zero factual hallucinations
- Support for multi-paragraph document inputs
- Sub-200ms latency on standard CPUs
02 / Standalone OS Harness Installers, CLI & Python SDK
In SLMAgents 2.0, SLM Summarizer 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_summarizer + slm_harness + the ./slm CLI), install the full 26-agent desktop suite, or install via Python pip:
1. Individual SLM Summarizer + 10-Plane Harness Installers (All OSes)
slm_summarizer + slm_harness + ./slm CLIslm_summarizer + slm_harnessslm_summarizer + 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-summarizer $ pip install -e ./slm_harness -e ./slm_summarizer # Option B: Run directly through the Unified 10-Plane Pure-CPU Harness CLI (./slm) $ ./slm ask "Summarize @quarterly_report.txt into 5 executive bullet points preserving all metrics" $ ./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 Summarizer 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 Summarizer on Pure CPU |
|---|---|---|
| Plane 6 | CascadeRouter (<35ms) |
Stage 2 (Summarization / TL;DR / Map-Reduce Intent) & Stage 3 (Embedding Margin) |
| Plane 2 & 5 | MemoryArena & Hybrid CPU Runtime |
Specialist Role: primary (Qwen3-4B GGUF Q4_K_M / Llama-3.2-3B INT4). 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 3 256-Token Chunked Prefill & Map-Reduce Hierarchical Folding + Key-Entity Retention Verifier |
| 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_summarizer_workspace && cd slm_summarizer_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_summarizer $ git sparse-checkout init --cone $ git sparse-checkout set slm_harness slm_summarizer # 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_summarizer or download the pre-packaged OS Harness Installer above.
05 / Configuration & API Reference
Constructor Parameters
Instantiate SLMSummarizer with performance and runtime options:
| Parameter | Type / Default | Description |
|---|---|---|
model_path | str | None | Local GGUF (.gguf) or quantized ONNX model path managed by MemoryArena. Default: None (auto-resolved via slm_harness). |
temperature | float | 0.3 | Sampling temperature. Default: 0.3. |
top_p | float | 0.9 | Nucleus sampling threshold. Default: 0.9. |
max_tokens | int | 512 | Max token output limit. Default: 512. |
n_threads | int | 4 | CPU threads for inference. Default: 4. |
Methods
| Method Signature | Return Type | Description |
|---|---|---|
summarize(text, ...) | str | Generates concise summary highlights from raw text documents. |
Execution Parameters
Complete list of execution parameters accepted by the primary agent method:
| Parameter | Type / Default | Description |
|---|---|---|
text | str | Raw text content string to summarize. |
system_prompt | str | None | Instruction defining target summary format. Default: None. |
user_input | str | None | Additional topic focus constraints. Default: None. |
temperature | float | 0.3 | Sampling temperature. Default: 0.3. |
top_p | float | 0.9 | Nucleus sampling probability. Default: 0.9. |
max_tokens | int | 512 | Maximum output token limit. Default: 512. |
Quick Start
from slm_summarizer import SLMSummarizer
summarizer = SLMSummarizer(temperature=0.3, top_p=0.9, max_tokens=512)
summary = summarizer.summarize(
text="Q3 net revenue reached $1.25M (+15% YoY). Operating margins expanded to 34% due to hardware optimization.",
system_prompt="Limit summary to concise bullet points",
user_input="Focus on revenue and margins",
temperature=0.3,
top_p=0.9,
max_tokens=512
)
print(summary)
06 / Verified Execution Logs
Diagnostic execution console output running locally on CPU:
→ INPUT: Text: Q3 net revenue reached $1.25M (+15% YoY)... ← OUTPUT: • Q3 net revenue grew +15% YoY to $1.25M • Operating margins expanded to 34% via local optimization