Base Model: Qwen3.5-27B BF16 with YaRN extension
Architecture: Qwen3.5 MoE
Context: 1M tokens (1,048,576)
Purpose: Long document analysis, book-length reasoning, extended chain-of-thought
This model uses YaRN (Yet another RoPE extensioN) for 1M context:
--ctx-size 1048576 --rope-scaling yarn --yarn-orig-ctx 262144
--temp 0.6 --top-p 0.95 --top-k 20 --min-p 0 --reasoning on
--temp 0.7 --top-p 0.8 --top-k 20 --min-p 0
llama-server -m Qwen3.5-27B-Opus-v2-YaRN-1M-Q4_K_M.gguf --ctx-size 1048576 --temp 0.6 --top-p 0.95 --top-k 20 --reasoning on --rope-scaling yarn --yarn-orig-ctx 262144
For standard 262K context, use the regular Opus v2 repo (smaller, faster).
Have the files? Submit a torrent.
Submit a torrentHave the complete files and permission to share them? Replace the example folder path. The script checks the files, makes a torrent and submits its magnet and checksums. Keep your torrent client seeding. Get your community key from your settings, and read the script before running it.
curl -fsSL https://pirateface.co/package.sh | bash -s -- --repo Exil01/Qwen3.5-27B-Opus-v2-YaRN-1M ./model-folder