Rocinante-12B-v1.1-GPTQ is a quantized version of a 12 billion parameter large language model optimized for efficient inference. Hosted on Hugging Face, it supports text generation and chat use cases through a standard chat template. It is designed for developers who want to run capable open-weight models locally or on modest hardware using the GPTQ quantization format and the Transformers library.
In the Foundation models & chat space, Rocinante 12B takes a focused approach. It focuses on accessing and running a high-quality open-source 12B language model locally without high compute requirements. It is built as an open-source project for developers. The project is open source (Open Source). Rocinante 12B is available on the web and API.
It is developed by AlfonsoM, and it first shipped in 2024. Key capabilities include quantized weights, GPTQ format, and chat template.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match