AI is happening. The future is updating.K2-Horizon-MoVA-36B-A4B-GGUF
> [!NOTE] > This repository contains GGUF versions of the IFM/K2-Horizon-MoVA-36B-A4B for use with `llama.cpp`. > > Multiple precision and quantization variants are available, including `BF16`, `Q8_0`, `Q6_K`, `Q5_K_M`, `Q5_0`, and `Q4_K_M`. All GGUF files include tokenizer metadata and a llama.cpp-compatible chat template. > [!WARNING] > **π‘ Important Deployment Notes:** These models require a version of `llama.cpp` containing K2 Horizon architecture support. Mainline support merged in release b11454. K2-Horizon-MoVA-36B-A4B is the sparse member of the K2-Horizon family: a Mixture-of-Experts model with Mixture-of-Values attention (MoVA) that stores 36B parameters and runs 4B per token. We have released the final checkpoint; intermediate checkpoints, along with the data and the training code, will be released.
> [!NOTE] > This repository contains GGUF versions of the IFM/K2-Horizon-MoVA-36B-A4B for use with `llama.cpp`. > > Multiple precision and quantization variants are available, including `BF16`, `Q8_0`, `Q6_K`, `Q5_K_M`, `Q5_0`, and `Q4_K_M`. All GGUF files include tokenizer metadata and a llama.cpp-compatible chat template. > [!WARNING] > **π‘ Important Deployment Notes:** These models require a version of `llama.cpp` containing K2 Horizon architecture support. Mainline support merged in release b11454. K2-Horizon-MoVA-36B-A4B is the sparse member of the K2-Horizon family: a Mixture-of-Experts model with Mixture-of-Values attention (MoVA) that stores 36B parameters and runs 4B per token. We have released the final checkpoint; intermediate checkpoints, along with the data and the training code, will be released.