# Jan-nano Technical Report

arXiv: https://arxiv.org/abs/2506.22760 · HTML: https://mlsanity.com/paper/2506.22760/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Jun 2025.

## Built on by (most cited first, 2 of 2)

- [A Survey of Reinforcement Learning for Large Reasoning Models](https://mlsanity.com/paper/2509.08827/) (2025)
- [Lucy: edgerunning agentic web search on mobile with machine generated task vectors](https://mlsanity.com/paper/2508.00360/) (2025)

## Abstract (arXiv)

Most language models face a fundamental tradeoff where powerful capabilities require substantial computational resources. We shatter this constraint with Jan-nano, a 4B parameter language model that redefines efficiency through radical specialization: instead of trying to know everything, it masters the art of finding anything instantly. Fine-tuned from Qwen3-4B using our novel multi-stage Reinforcement Learning with Verifiable Rewards (RLVR) system that completely eliminates reliance on next token prediction training (SFT), Jan-nano achieves 83.2% on SimpleQA benchmark with MCP integration while running on consumer hardware. With 128K context length, Jan-nano proves that intelligence isn't about scale, it's about strategy.