Tools to convert/quantise language models in GGUF format
Updated 2025-08-09 17:16:09 +01:00
Self-hosted, GPU-optimised GenAI platform providing a drop-in OpenAI-compatible API
Updated 2025-08-06 10:20:27 +01:00
MCP server to provide OpenAPI-flavour tools to an Open WebUI instance
Updated 2025-08-01 21:39:39 +01:00
Script and results for Nvidia GPU benchmarking across various context lengths
Updated 2025-07-28 21:21:56 +01:00
Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.
Updated 2025-07-14 18:12:33 +01:00