Released on Sep 25, 2026
27B reasoning model with ternary weights at 1.72 bits per weight, derived from Qwen3.8-27B with hybrid attention (~75% linear). In GGUF PTQ1_0 it takes 5.95 GB, retains 98.2% of the FP16 intelligence and reaches ~47 tok/s on an Apple M5 Max laptop, with a 262K-token context. Apache 2.0 license. Requires Prism ML's llama.cpp fork (ternary kernels are not in stock llama.cpp).
Modello di ragionamento 27B ternario (g128, 1,72 bit/peso) derivato da Qwen3.8-27B con attenzione ibrida (~75% lineare): 5,95 GB in PTQ1_0, contesto 262K, 98,2% dell'intelligenza FP16, ~47 tok/s su M5 Max. Richiede la fork llama.cpp di Prism ML.
Visit the official site to download, learn more and start using Bonsai-2 27B.
DeepSeek's MoE model with 552B total parameters and 8-16B active per token. 1M-token context, MIT license: free commercial use. Built for online serving, too large for local use on a single PC.
The most advanced open image model: excellent prompt adherence, solid anatomy, full fine-tuning. The standard for serious work.
Very few steps, high quality, and strong text rendering. The distilled FLUX variant, built for speed.
