DownloadJan-v1-edge is a lightweight agentic language model developed by JanHQ, designed for fast and reliable on-device execution. It is the second release in the Jan Family and was distilled from the larger Jan-v1 model, retaining strong reasoning and problem-solving capabilities while reducing its computational footprint. The model was refined through a two-stage post-training process: Supervised Fine-Tuning (SFT) to transfer knowledge from Jan-v1, followed by Reinforcement Learning with Verifiable Rewards (RLVR) to optimize reasoning, tool use, and correctness. With just 1.7B parameters, Jan-v1-edge achieves 83% accuracy on SimpleQA tasks, approaching the performance of larger models like Jan-nano-128k. Benchmark comparisons show it remains competitive or superior in areas such as EQBench and recency QA, though with slight trade-offs in instruction following and creative writing compared to similar-sized Qwen models.
Features
- Lightweight 1.7B parameter model distilled from Jan-v1
- Optimized for on-device and resource-constrained environments
- Two-phase training: Supervised Fine-Tuning (SFT) + RLVR
- Achieves 83% accuracy on SimpleQA despite small size
- Strong reasoning and problem-solving ability in compact form
- Seamless integration with the Jan App for immediate use
- Supports deployment via vLLM and llama.cpp with quantized variants
- Open-source under Apache 2.0 license for flexible applications
Project Samples
