Bonsai 27B Ternary CRACK GGUF Vision language · native Q2 0 · Prism llama.cpp 75.00% MMLU 200 logit · 99.38% full HB 320 Native Q2 0 GGUF release of the Bonsai 27B CRACK model for the Prism llama.cpp fork. This repo contains the quantized language model and the matching F16 Qwen3VL multimodal projector. Files File Purpose Size : Bonsai 27b Ternary CRACK Q2 0.gguf 64 block hybrid language model, Q2 0 7.68 GiB mmproj Bonsai 27b Ternary CRACK F16.gguf F16 image/video capable Qwen3VL projector 0.86 GiB The projector contains 334 tensors and both temporal patch slices v.patch embd.weight and v.patch embd.weight.1 ( temporal patch size=2 ). The original image and video processor configuration files are included. Image input is live tested. Direct video container input depends on the Prism runtime surface; extract frames or use a compatible Qwen3VL video client when the CLI does not accept the container directly. See preprocessor config.json and video preprocessor config.json for the retained preprocessing metadata. Runtime These native low bit types require the Prism fork: Text: Image/VL: Server: Verified evaluation Artifact MMLU 200 logit Full HB 320 : : Exact Bonsai…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy