Morning everyone! I put this merge together yesterday, put it on huggingface to share with Jackrong, and a bunch of people discovered it whe…

Morning everyone! I put this merge together yesterday, put it on huggingface to share with Jackrong, and a bunch of people discovered it when it got cloned into Jackrongs repo, and it’s going a bit viral. But it’s completely not ready lol so hang tight. I am currently doing a fine tune run in hopes of smoothing the transition between the model layers and improve the hallucination and looping. Regardless, this run is just a fun experiment, fine tuning is running for me and has 165 samples left. If I can get it cleaned up, it will be really nice for people with 12-16GB GPUs which was the goal with this project. The new one will be uploaded here to my repo when complete, no guarantees, we’re just having fun and experimenting here! Also there’s a bunch of people watching Jackrongs repos to repost his new models first. The issue is we’ve been using hugging face to send completely beta models back and forth, so people keep posting them when they are totally not ready yet. If I post about a model release, it’s because Jackrong told me it’s ready, and I’ve already ran the benches! Other than this one, but it might be ready to play with soon: https://huggingface.co/KyleHessling1/Qwopus-GLM-18B-Merged-GGUF

Related

• We absolutely LOVE local models. Have been talking about this for years, it is my favorite thing. Most people cannot afford luxury AI tokens…
• So excited to share that @adrgrondin and @LocallyAIApp are joining LM Studio family! Together we are doubling down on Apple platforms to bri…
• Today, we’re excited to announce that we’re joining @lmstudio! We share a common vision of building amazing products for everyone and making…
• .@steipete has been an amazing supporter of open models and local models. This is even before the rise of OpenClaw. 🦞 Ollama is here to lov…

Source: Clem Delangue (X) | 2026-04-18