PANews reported on December 2nd that Tether Data announced the release of QVAC Fabric LLM, a new comprehensive runtime environment and fine-tuning framework for large language model (LLM) inference. This framework supports running, training, and customizing large language models directly on everyday hardware such as consumer GPUs, laptops, and even smartphones. Tasks that previously required high-end cloud servers or dedicated NVIDIA systems can now be accomplished locally on users' existing devices.

QVAC Fabric LLM also expands the capabilities of the llama.cpp ecosystem by adding fine-tuning support for modern models such as LLama3, Qwen3, and Gemma3. Supporting training on a wide range of GPUs, including AMD, Intel, NVIDIA, Apple chips, and mobile chips, QVAC Fabric LLM breaks the long-held assumption that meaningful AI development requires specialized hardware from a single vendor. Tether Data has released QVAC Fabric LLM as open-source software under the Apache 2.0 license and provides multi-platform binaries and ready-to-use adapters on Hugging Face. Developers can begin fine-tuning with just a few commands, lowering the barrier to AI customization.