cd /news/ai-agents/system-1-agents-a-method-for-local-r… · home › topics › ai-agents › article
[ARTICLE · art-173736] src=discuss.huggingface.co ↗ pub= topic=ai-agents verified=true sentiment=↑ positive

System 1 agents: a method for local, reliable and fast agentic AI

A new preprint on Hugging Face introduces "System 1 agents," local AI agents that never generate text and therefore cannot hallucinate by design, according to the paper's authors. Tested on 1,952 requests to 157 medical imaging and biomedical tools across 18 MCP servers, six System 1 agents achieved 97.0% correct tool choice with zero invented values in 10,888 calls, versus 5.5–19.7% of calls for nine LLM-based agents, and were 92.9% fully correct at 2.6 s and 0.228 Wh per correct request against 91.3%, 4.5 s and 0.413 Wh for the best LLM agent. The method uses a deterministic compiler that turns each request and tool schema into multiple-choice questions answered by a decision model, and runs entirely on one local GPU so sensitive data stays inside the institution.

recommendation

Replace free-text tool selection with a deterministic compiler that turns requests and schemas into multiple-choice questions, so the agent cannot invent values or fabricate missing parameters.

more in playbook: Designing agent systems
read1 min views6 publishedNov 21, 2026

Hey all, we just released a methodology to develop local, reliable and fast System 1 agents!

What if an AI agent could call tools without writing a single word?

Our new preprint introduces System 1 agents: local AI agents that cannot hallucinate by design, because they never generate text. A deterministic compiler turns each request and the tool’s schema into multiple-choice questions, and a decision model answers them by choosing among options that come only from the request, the schema and the tool catalogue.

We developed six System 1 agents and compared them with nine LLM-based agents on 1,952 requests to 157 medical imaging and biomedical tools from 18 MCP servers:

→ 97.0% correct tool choice, significantly ahead of all local LLM-agent settings

→ Zero invented values in 10,888 calls, against 5.5–19.7% of LLM-agent calls

→ Missing information? It asks the user. LLM agents made it up in up to 62.1% of such requests

→ More accurate and cheaper: 92.9% fully correct at 2.6 s and 0.228 Wh per correct request, against 91.3%, 4.5 s and 0.413 Wh for the best LLM agent

→ Everything runs on one local GPU, so sensitive data never leave the institution

 Paper page: [Paper page - System 1 agents: a method for local, reliable and fast agentic AI](https://huggingface.co/papers/2611.21053)

 Code and data: [GitHub - eltzanis/SystemOneAgents: A method to develop System 1 Agents · GitHub](https://github.com/eltzanis/SystemOneAgents)

Feedback and questions are very welcome!

── more in #ai-agents 4 stories · sorted by recency
── more on @hugging face 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/system-1-agents-a-me…] indexed:0 read:1min 2026-11-21 · —