Slipstream

Semantic Quantization for Multi-Agent AI Communication

---

82% fewer tokens. Factorized Force-Object intents. Built for the AAIF ecosystem.

Before (45 tokens):
{"from": "alice", "to": "bob", "type": "request", "action": "review", "target": "auth_module"}

After (6 tokens):
SLIP v3 alice bob Request Review auth

Multi-agent AI systems waste 40-60% of compute on coordination overhead. At scale, that's $180K-$2.5M/year just for agents talking to each other.

Slipstream fixes this through semantic quantization - transmitting factorized intents (Force + Object) rather than verbose messages.

v3 Innovation: Factorized 2-token intents replace ~46 flat anchors. SLIP v3 src dst Request Plan instead of SLIP v1 src dst RequestPlan. This reduces the classification problem from 46-way to 12-way + 30-way, making it learnable by small models.

---

Quick Start

pip install slipcore

from slipcore import format_slip, parse_slip, render_human, KeywordQuantizer

# Create a message (6 tokens instead of 45)
wire = format_slip("alice", "bob", "Request", "Review", ["auth"])
# -> "SLIP v3 alice bob Request Review auth"

# Or let the quantizer map natural language
q = KeywordQuantizer()
wire = q.quantize(
    "Please check the authentication code for security issues",
    src="dev", dst="reviewer"
)
# -> "SLIP v3 dev reviewer Request Review"

# Parse
msg = parse_slip(wire)
print(msg.force, msg.obj, msg.payload)
# Request Review ['auth']

# Human-readable
print(render_human(msg))
# [alice -> bob] Request Review: "Request review of work" (payload: auth)

---

Why Slipstream?

The Problem

BPE tokenizers fragment compressed formats, negating syntactic optimization:

Compressed: REQ/TSK|s=7|d=3|act=review
Expected:   8 tokens
Actual:     22 tokens (every | and = is a token!)

The Solution

Slipstream uses a Universal Concept Reference (UCR) - a shared semantic manifold where common agent intents have factorized names (Force + Object) that tokenize efficiently across all LLM architectures.

Format	Tokens	Annual Cost (50 agents)
JSON verbose	~45	$180,000
JSON minimal	~30	$120,000
Slipstream v3	~6-8	$32,000

---

Wire Format

SLIP v3 <src> <dst> <Force> <Object> [payload...]

• Factorized intents - Force (action verb) + Object (domain noun)
• No special characters - avoids BPE fragmentation
• Space-separated - clean tokenization
• 12 Force tokens - closed vocabulary, easily learned
• Zero core dependencies - stdlib-only core package

Force Tokens (12 closed vocabulary)

Force	Description
Observe	Passively notice state/change/error
Inform	Report information (status, completion, blockage)
Ask	Request information (clarification, status, permission)
Request	Ask for action (task, review, help, plan)
Propose	Suggest something (plan, change, alternative)
Commit	Commit to something (task, deadline, resource)
Eval	Evaluate work (approve, needs work)
Meta	Protocol-level (acknowledge, sync, handoff)
Accept	Accept a proposal/request
Reject	Decline a proposal/request
Error	Report system error
Fallback	Content too specific for standard tokens

Core Object Tokens

Task, Plan, Review, Help, Status, Complete, Blocked, Progress, State, Change, Error, Result, Clarify, Permission, Resource, Cancel, Priority, Alternative, Rollback, Deadline, Approve, NeedsWork, Ack, Sync, Handoff, Escalate, Abort, Condition, Defer, Timeout, Validation, Generic

---

Finetuned Model

We provide a ready-to-use model trained on the Slipstream protocol:

Format	Link	Use Case
LoRA Adapter	slipstream-glm-z1-9b	Merge with base
GGUF Q4	slipstream-glm-z1-9b-gguf	Ollama / llama.cpp
Dataset	slipstream-tqt	Train your own

Run with Ollama

ollama run anthony-maio/slipstream

Train Your Own

# Generate v3 training dataset
python -m slipcore.finetune -n 1000 -f sharegpt_thought -o train.jsonl

# Or use LLM-enhanced generation
python -m slipcore.finetune_llm -n 1000 --provider gemini -o train.jsonl

# Migrate existing v2 data to v3
python scripts/migrate_v2_data.py data/slipstream-tqt.jsonl data/slipstream-tqt-v3.jsonl

---

AAIF Integration

Slipstream is designed as the transport layer for the Linux Foundation Agentic AI ecosystem:

+-------------------------------------+
|   Application (Agent Logic)         |
+----------------+--------------------+
                 |
+----------------v--------------------+
|   MCP / A2A (Semantic Layer)        |
+----------------+--------------------+
                 |
+----------------v--------------------+
|   Slipstream (Transport Layer)      |  <- 82% token reduction
+----------------+--------------------+
                 |
+----------------v--------------------+
|   Network                           |
+-------------------------------------+

---

Resources

• Paper: Slipstream: Semantic Quantization for Efficient Multi-Agent Coordination
• Model: HuggingFace
• Dataset: HuggingFace
• Spec: spec/spec-00-invariants.md

---

Citation

@misc{maio2025slipstream,
  title={Slipstream: Semantic Quantization for Efficient Multi-Agent Coordination},
  author={Maio, Anthony},
  year={2025},
  url={https://github.com/anthony-maio/slipcore}
}

---

License

Apache 2.0

---

Stop paying the token tax.

pip install slipcore