pre-reasoning 2.5.0

3M-parameter neural pre-reasoning engine for grounding LLMs before they answer.

Pre-Reasoning

Pre-Reasoning is a Mia Labs structural analysis engine that grounds an LLM before it answers. It uses a 3M-parameter neural model (V3) to surface dependencies, root blockers, unlock order, parallel work, cycles, and conflicts from problem text.

The engine ships with bundled safetensors weights -- install and run, no downloads needed.

What It Does

Given natural-language problem text, the engine returns:

• ROOT BLOCKERS: what must be resolved first
• UNLOCK SEQUENCE: a dependency-aware resolution order
• PARALLEL WORK: independent items that can proceed now
• CYCLES: circular dependencies that cannot be solved sequentially
• CONFLICTS: competing positions or incompatible entities
• REQUIREMENTS: numeric or threshold requirements

When torch is not installed, the engine falls back to deterministic heuristic graph reasoning. For the best experience, install the [neural] extra.

Install

pip install pre-reasoning              # base (deterministic fallback)
pip install "pre-reasoning[neural]"    # full: 3M-param V3 neural model

For local development from this repo:

pip install -e .
pip install -e ".[neural]"

Python Usage

from pre_reasoning import analyze, pulse

result = analyze("Frontend depends on API. API depends on Auth.")
print(result["trace"])

check = pulse(
    "Frontend depends on API. API depends on Auth.",
    "Fix Auth first, then verify the API before frontend work."
)
print(check["status"])

CLI Usage

pre-reasoning "A depends on B. B depends on C."
pre-reasoning --json "CTO conflicts with senior dev."
pre-reasoning --info

To use a different weights file, set PRE_REASONING_CHECKPOINT=/path/to/weights.safetensors or pass --checkpoint.

Results

Early comparison table, illustrative, n=5 architectural decision problems:

Comparison	Illustrative result, n=5
9B + trace vs 32B baseline	3W 2T 0L
9B + trace vs 120B baseline	4W 1T 0L
120B + trace vs 120B baseline	3W 2T 0L

These are product-research notes, not benchmark claims.

Architecture

Full install (recommended):
User text
  -> V3 neural perception (3M params, safetensors)
  -> neural findings converted to structural blocks
  -> heuristic graph analysis
  -> structural trace

Base install (fallback when torch is missing):
User text
  -> deterministic text-to-blocks adapter
  -> heuristic graph analysis
  -> structural trace

File Map

Path	Purpose
pre_reasoning/	Installable Python package and CLI entry point
pre_reasoning/inference.py	3M-parameter V3 neural perception layer
pre_reasoning/heuristic.py	Deterministic graph-reasoning core (fallback)
pre_reasoning/pre_reasoning_v2_5.py	v2.5 orchestrator: V3 neural + heuristic
pre_reasoning/checkpoints/pre-reasoning-3m-v2.5.safetensors	Bundled V3 weights (11MB)
examples/	Runnable usage examples
tests/	Pytest suite
skill/SKILL.md	Agent skill descriptor for model adoption
CLAUDE.md	Optional Claude Code hooks configuration
WHY_TRACES_WORK.md	Literature connection, 9 cited papers

Weights Policy

The raw training checkpoint is not part of the release. The package bundles pre_reasoning/checkpoints/pre-reasoning-3m-v2.5.safetensors, a weights-only inference artifact. It ships no training metadata: no optimizer state, LR schedules, step counters, RNG state, training config, or raw checkpoint provenance.

License

MIT License. See LICENSE.

Authors

Luis Lozano and Dr. Shannon, Mia Labs' AI co-researcher, 2026.