Jan-Nano-128k represents a significant advancement in compact language models for research applications. Building upon the success of
Jan-Nano, this enhanced version features a
native 128k context window that enables deeper, more comprehensive research capabilities without the performance degradation typically associated with context extension methods.
This model maintains full compatibility with Model Context Protocol (MCP) servers while dramatically expanding the scope of research tasks it can handle in a single session.
Jan-Nano-128k has been rigorously evaluated on the SimpleQA benchmark using our MCP-based methodology, demonstrating superior performance compared to its predecessor:
Traditional approaches to extending context length, such as YaRN (Yet another RoPE extensioN), often result in performance degradation as context length increases. Jan-Nano-128k breaks this paradigm:
This fundamental difference makes Jan-Nano-128k ideal for research applications requiring deep document analysis, multi-document synthesis, and complex reasoning over large information sets.
Jan-Nano-128k is fully supported by
Jan - beta build, providing a seamless local AI experience with complete privacy and control.
For additional tutorials and community guidance, visit our
Discussion Forums.
1vllm serve Menlo/Jan-nano-128k \
2 --host 0.0.0.0 \
3 --port 1234 \
4 --enable-auto-tool-choice \
5 --tool-call-parser hermes \
6 --rope-scaling '{"rope_type":"yarn","factor":3.2,"original_max_position_embeddings":40960}' --max-model-len 131072
Note: The chat template is included in the tokenizer. For troubleshooting, download the
Non-think chat template.
1Temperature: 0.7
2Top-p: 0.8
3Top-k: 20
4Min-p: 0.0
1@model{jan-nano-128k,
2 title={Jan-Nano-128k: Deep Research with Extended Context},
3 author={Dao, Alan and Dinh, Bach Vu and Le Thinh},
4 year={2024},
5 url={https://huggingface.co/Menlo/Jan-nano-128k}
6}
