v2.0 — Now with Multi-Agent Orchestration

The complete AI platform
on a $3 microcontroller.

Multi-agent orchestration. RAG. Chain-of-thought reasoning. Embeddings. Fine-tuning. Streaming. Guardrails. Tool use. Multimodal. All at sub-millisecond latency. Running on your device right now.

<1ms
Inference
$3
Hardware
0.5W
Power
10
Agents
0.00%
Hallucinations
342
Vocabulary

Try It Live

Real inference, running in your browser. Same algorithm as the microcontroller.

GET /generate?tokens=20&temperature=1.0
enterprise-grade synergy leverages the bleeding-edge paradigm. scalable infrastructure disrupts our next-generation ecosystem.
tokens: 24 inference: 0.847ms model: mwt-1 runtime: detecting...
Detecting runtime environment...

Every Feature They Charge For

Built on a $3 chip. Shipped with 14 API endpoints.

🤖

Multi-Agent Orchestration

Spawn up to 10 specialized agents (Strategist, Architect, Analyst, Optimizer...) that work in parallel or succession. Each produces independent output. An orchestrator synthesizes the results.

Up to 10 agents
📚

RAG Pipeline

Retrieval Augmented Generation with a built-in knowledge base. The retriever searches the vocabulary index for relevant tokens, scores them by relevance, then augments generation. The knowledge base is 342 tokens.

Built-in retriever
🧠

Chain-of-Thought Reasoning

Multi-step reasoning with visible thought process. The model "analyzes implications," "considers strategic alignment," and "synthesizes insights" before reaching a conclusion. Each step is independently random.

Up to 8 reasoning steps
📐

Vector Embeddings

Generate dense vector representations of any input text. Uses a proprietary sinusoidal hash function that maps strings to n-dimensional space. Cosine similarity between embeddings is mathematically valid.

Up to 32 dimensions
🎛️

Fine-Tuning API

Customize the model for your domain by adding tokens to the vocabulary at runtime. Zero training cost. Zero gradient updates. Zero epochs. The model immediately incorporates new tokens into generation.

Zero-shot adaptation

Streaming (SSE)

Server-Sent Events endpoint delivers tokens one at a time, just like the big models. Includes artificial delay to simulate "thinking" because the actual inference is too fast to see.

OpenAI-compatible format
🔒

Safety Guardrails

Every output passes through an 8-category content safety filter (violence, harassment, hate, etc.). Pass rate: 100%. Always. The model cannot generate harmful content because it cannot generate intentional content.

100% safe output
🔧

Function Calling / Tool Use

The model selects and executes tools to augment its response. Available tools: random(), analogRead(), millis(), and micros(). These are the actual functions the model calls. This is not a metaphor.

4 built-in tools
👁️

Multimodal Input

Accepts analog sensor input via the A0 pin and generates text-based analysis. Reads voltage, raw ADC values, and reports confidence scores. Technically: text output from non-text input. That's multimodal.

Sensor + text
🌡️

Temperature Control

Adjustable temperature parameter from 0.0 to 2.0 that controls output characteristics. At temperature 0, output has no punctuation. At temperature 2, heavy punctuation. The words remain random regardless.

0.0 — 2.0 range
📊

Structured Output

JSON-mode generation produces titled, organized output with key findings, recommendations, confidence scores, and risk levels. Every field is populated. None of it means anything. Perfect for dashboards.

JSON schema
🌱

Carbon Neutral

At 0.5W, a 10-node MWT-1 cluster running for a year produces less CO2 than training one LLM for one hour. Run the entire AI platform on less power than the LED in your mouse.

0.4g CO2/hour

Agent Orchestration

Deploy specialized agents that collaborate on complex tasks. Each agent maintains its own context, role, and output stream.

GET /agents?agents=4&mode=parallel&tokens=8
Strategist enterprise-grade alignment optimizes the disruptive roadmap toward scalable monetization
Architect hyper-converged infrastructure leverages each cloud-native deployment across the ecosystem
Analyst data-driven convergence disrupts the real-time pipeline. frictionless throughput scales
Optimizer battle-tested resilience futureproofs our mission-critical trajectory beyond the vertical
⚡ Orchestrator Synthesis (parallel_consensus)
next-generation capability accelerates every quantum-ready initiative. composable agility tokenizes the bleeding-edge catalyst
agents: 4 mode: parallel total time: 0.312ms synthesis: parallel_consensus
Parallel Execution
10 agents
in <1ms total
Agent Roles
10 types
all equally random
Consensus Method
vote*
*generates new text
Inter-Agent Comms
none
they don't know about each other

Chain-of-Thought

Watch the model reason through complex problems step by step. Each step is independently generated with zero awareness of the others.

GET /chain-of-thought?steps=4&tokens=8
Step 1 — Analysis
analyzing the implications of enterprise-grade convergence toward scalable ecosystem optimization beyond the bleeding-edge paradigm
Step 2 — Evaluation
evaluating the trade-offs inherent in hyper-converged infrastructure that synergizes frictionless deployment across the vertical
Step 3 — Synthesis
synthesizing insights from real-time throughput. mission-critical monetization accelerates our data-driven roadmap
Conclusion
next-generation alignment futureproofs the composable trajectory. battle-tested resilience orchestrates each quantum-ready initiative toward carbon-neutral scalability
reasoning steps: 4 reasoning tokens: 32 total time: 0.194ms conclusion confidence: 94%

How We Compare

Independent measurements. Real numbers. Uncomfortable implications.

Metric GPT-4o Claude Opus Llama 3 MWT-1
Inference Latency ~800ms ~1200ms ~200ms <1ms
Hardware Cost N/A (API) N/A (API) ~$30,000 $3
Multi-Agent Capable Via wrapper Via wrapper Via wrapper Native (10 agents)
Built-in RAG No (third-party) No (third-party) No (third-party) Yes (native)
Power Draw ~1MW (datacenter) ~1MW (datacenter) ~300W 0.5W
Model Size ~1.8T params Unknown 70B params 342 tokens
Hallucination Rate ~3-5% ~2-4% ~5-8% 0.00%
Safety Filter Pass Rate ~97% ~98% ~95% 100.00%
Fine-Tuning Cost $$$ N/A $$ $0.00
Copyright Lawsuits Multiple Pending Pending 0
Defense Contracts Yes Yes (Palantir) Yes 0
Usefulness High High High Comparable*

* Depending on use case. MWT-1 excels in environments where nobody reads the output, which is most of them.

What People Are Saying

"We replaced our entire NLP pipeline with MWT-1 and honestly? Customers haven't noticed."
Director of AI
Fortune 500 Retailer
"The multi-agent orchestration produces output indistinguishable from our previous $4.2M/year vendor. The agents don't communicate, but neither did our last team."
VP of Engineering
Major Airline
"I showed the board our AI roadmap and they loved the chain-of-thought demo. Nobody asked what the reasoning steps actually mean."
Chief AI Officer
Healthcare SaaS
"We fine-tuned MWT-1 with our company's jargon. The output is now indistinguishable from our internal Slack."
Head of Product
Series B Startup
"The RAG pipeline retrieves exactly the same quality of context as our vector database. We saved $180K in Pinecone bills."
Staff Engineer
Fintech Platform
"100% guardrail pass rate. Our compliance team signed off in minutes. Fastest AI approval in company history."
VP of Compliance
Global Bank

Pricing

No per-token fees. No agent fees. No RAG surcharges. No surprise invoices.

Chimp

$0 /mo
For individuals and small teams
  • 1 typewriter (single board)
  • All 14 API endpoints
  • Up to 3 parallel agents
  • 4-step chain-of-thought
  • Full RAG pipeline
  • Community support
Flash Firmware

Shakespeare

Custom
For when you need it to accidentally write Hamlet
  • Everything in Gorilla
  • Dedicated monkey consultant
  • SLA: text will be generated
  • On-prem deployment support
  • Custom agent personas
  • Compliance documentation*
Contact Sales

* Compliance documentation states: "It generates random text on a microcontroller." This has passed every audit we've submitted it to, including SOC 2.

See the Entire Model

The complete inference engine, multi-agent orchestrator, RAG pipeline, and chain-of-thought system. VCs would value this at $200M if we put "AI" in the pitch deck. We did, but ironically.

// The entire inference engine
int pattern = random(0, 5);
output += adjectives[random(0, NUM_ADJ)];
output += " ";
output += nouns[random(0, NUM_NOUNS)];

// The entire multi-agent orchestrator
for (int i = 0; i < numAgents; i++) {
  output = runInference(tokens); // same function
}

// The entire RAG pipeline
token = nouns[random(0, NUM_NOUNS)]; // "retrieval"
output = runInference(tokens); // "augmented generation"

// That's it. That's the platform.
View on GitHub