Provides GGUF-format fine-tuned Qwen3.6-27B weights optimized for consumer hardware, offering NEO IMATRIX and MTP quant variants, vision support, 256k native context, and uncensored 'heretic' traces with published benchmark improvements over the base model.
[next-mdx-remote-client] error compiling MDX: Unexpected character `5` (U+0035) before name, expected a character that can start a name, such as a letter, `$`, or `_` More information: https://mdxjs.com/docs/troubleshooting-mdx
Provides a locally deployable, FP4-quantized variant of Qwen3.8-Flash-Next for multimodal (text, image, video) text generation. Uses NVIDIA Model Optimizer with quantization-aware distillation (QAD), a ~96 GiB checkpoint, and is validated on vLLM for Blackwell and DGX Spark.
Generates images from natural-language prompts or structured JSON layouts, emphasizing readable long-form bilingual text and precise layout control. Built on a compact 5B diffusion backbone, supports prompts up to 4K tokens and ships as a Diffusers pipeline under Apache‑2.0.
Provides a GGUF-quantized build of the LFM2.5 24B agent-oriented model for on-device CPU inference with llama.cpp, enabling local text generation and tool-calling workflows; includes a QAD Q4_0 checkpoint and recommended sampling defaults.
