Three models. One decision. Inspired by the MAGI supercomputer from Neon Genesis Evangelion.
MAGI is not another agent framework. It is a structured disagreement engine: the same question goes to three different LLMs, each with a different perspective. They vote, debate, and critique each other to produce a Decision Dossier with the ruling, confidence, minority report, and full trace.
"Three cobblers with their wits combined equal Zhuge Liang, the master mind."
In our latest MMLU (Massive Multitask Language Understanding) "Hell Mode" benchmark:
- Single Strong Model (Claude Sonnet 4.6): 83.3%
- MAGI Critique (3x Cheap Models*): 83.3% 🤝 (MATCHED)
Models: Xiaomi Mimo-v2-pro, MiniMax-m2.7, DeepSeek-v3.2. Note: MAGI achieves the same accuracy as the world's leading model by leveraging collective intelligence through multi-round debate.
The value: MAGI allows you to achieve State-of-the-Art (SOTA) performance using significantly smaller/cheaper nodes by enabling them to catch each other's logic gaps and hallucinations.
There are several EVA-inspired multi-model projects. Here's what makes this one different.
Other projects do voting. Three models answer, pick the majority. That's it.
MAGI does structured disagreement. Models don't just answer in parallel. They read each other's answers, critique the reasoning, and revise their positions across multiple rounds. The system tracks who changed their mind and why.
| Capability | Voting projects | MAGI |
|---|---|---|
| Multi-model query | Yes | Yes |
| Majority vote | Yes | Yes |
| Multi-round critique (ICE) | No | Yes |
| Mind change tracking | No | Yes |
| Adaptive protocol selection | No | Yes |
| Minority report / dissent analysis | No | Yes |
| Benchmark: ensemble >= single model | No | Yes (83% vs 83%) |
| Fault tolerance (node failures) | No | Yes |
| NERV hexagonal dashboard | No | Yes |
| CLI toolchain (diff, judge, bench) | No | Yes |
The key finding: MAGI enables cheap models to reach the "Sonnet Ceiling". While single strong models are elite, MAGI's critique mode allows an ensemble of much cheaper models to reach the same level of accuracy (83.3%) by catching mistakes in high-stakes reasoning tasks.
pip install magi-systemOr from source:
git clone https://github.com/fshiori/magi.git
cd magi
uv venv && uv pip install -e ".[dev]"# Set your API key (OpenRouter gives you access to all models with one key)
export OPENROUTER_API_KEY=sk-or-...
# Ask a question — three models debate, one decision emerges
magi ask "Should we use microservices or a monolith?"
# Multi-model code review (the killer use case)
magi diff --staged
# Critique mode: models debate until consensus (slower, higher quality)
magi ask "Is Rust better than Go for backend services?" --mode critique
# Adaptive mode: auto-selects vote/critique/escalate based on disagreement
magi ask "What caused the 2008 financial crisis?" --mode adaptive
# Multi-model answer scoring
magi judge -q "What is quantum entanglement?" -a "It means particles are connected"
# NERV Command Center — real-time dashboard
pip install magi-system[web]
magi dashboard
# Run benchmark, view analytics, replay decisions
magi bench --dataset mmlu:abstract_algebra --use-judge
magi analytics
magi replay <trace-id>
# List persona presets
magi presets| Protocol | When to use | How it works |
|---|---|---|
vote |
Fast answers, clear-cut questions | Parallel query, structured position extraction, majority wins |
critique |
Complex or controversial questions | Multi-round debate (ICE), models critique each other until consensus |
escalate |
Forced decision on high-disagreement topics | Critique with 2-round limit, highest-trust node makes final call |
adaptive |
Default for most use cases | Auto-selects based on agreement score: high=vote, medium=critique, low=escalate |
MAGI comes with 5 built-in perspective sets:
$ magi presets
code-review Security Analyst / Performance Engineer / Code Quality Reviewer
eva Melchior / Balthasar / Casper
research Methodologist / Domain Expert / Devil's Advocate
strategy Optimist / Pessimist / Pragmatist
writing Editor / Reader Advocate / Fact Checker
Tested on MMLU (Massive Multitask Language Understanding) "Hell Mode" (Abstract Algebra, Professional Law, Formal Logic):
| Group | Accuracy | Strategy | Verdict |
|---|---|---|---|
| Claude Sonnet 4.6 (Single) | 83.3% | Single Shot | Peak individual performance |
| MAGI Critique (3x Cheap) | 83.3% | ICE Protocol | Matched. Beats Sonnet on Logic |
Models used: Xiaomi MiMo-v2-pro, MiniMax M2.7, DeepSeek V3.2. Judge: Verified by Gemini 3.1 Pro via OpenRouter.
MAGI keeps working when models fail:
- 1 of 3 fails — continues with 2 nodes, marks decision as degraded
- 2 of 3 fail — falls back to single model response
- All 3 fail — raises
MagiUnavailableError(never guesses) - Timeouts — 60s default per node, exponential backoff on rate limits
- Reasoning models — automatically extracts from
reasoning_content(e.g., MiniMax M2.7)
magi/
├── core/
│ ├── engine.py # MAGI engine, coordinates nodes
│ ├── node.py # LLM node wrapper with persona
│ └── decision.py # Decision dossier dataclass
├── protocols/
│ ├── vote.py # Structured voting with position extraction
│ ├── critique.py # ICE (Iterative Consensus Ensemble)
│ └── adaptive.py # Dynamic protocol selection
├── commands/
│ ├── diff.py # Multi-model code review
│ ├── judge.py # Multi-model answer scoring
│ └── analytics.py # Trace analysis and replay
├── web/
│ ├── server.py # FastAPI + WebSocket server
│ └── static/ # NERV Command Center UI
├── presets/ # Persona preset definitions
├── bench/ # Benchmark runner and datasets
├── trace/ # JSONL trace logging
└── cli.py # Click CLI entry point
In Evangelion, MAGI is a trio of supercomputers created by Dr. Naoko Akagi. Each embodies a different aspect of her personality: Melchior (the scientist), Balthasar (the mother), and Casper (the woman). Decisions are made by majority vote among the three.
MAGI applies this concept to LLMs: same question, three different perspectives, structured disagreement produces better decisions than any single model alone.
MIT