From 09346bdc069de7dc47ae29627e9a7103fcad9a42 Mon Sep 17 00:00:00 2001 From: Ettore Di Giacinto Date: Mon, 6 Oct 2025 08:46:09 +0200 Subject: [PATCH] chore(model gallery): add lemon07r_vellummini-0.1-qwen3-14b (#6386) Signed-off-by: Ettore Di Giacinto --- gallery/index.yaml | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/gallery/index.yaml b/gallery/index.yaml index 60402eee4..36dbd4740 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -2755,6 +2755,24 @@ - filename: aquif-3.5-A4B-Think.Q4_K_M.gguf sha256: 1650b72ae1acf12b45a702f2ff5f47205552e494f0d910e81cbe40dfba55a6b9 uri: huggingface://QuantFactory/aquif-3.5-A4B-Think-GGUF/aquif-3.5-A4B-Think.Q4_K_M.gguf +- !!merge <<: *qwen3 + name: "lemon07r_vellummini-0.1-qwen3-14b" + urls: + - https://huggingface.co/lemon07r/VellumMini-0.1-Qwen3-14B + - https://huggingface.co/bartowski/lemon07r_VellumMini-0.1-Qwen3-14B-GGUF + description: | + Just a sneak peek of what I'm cooking in a little project called Vellum. This model was made to evaluate the quality of the CreativeGPT dataset, and how well Qwen3 trains on it. This is just one of many datasets that the final model will be trained on (which will also be using a different base model). + + This got pretty good results compared to the regular instruct in my testing so thought I would share. I trained for 3 epochs, but both checkpoints at 2 epoch and 3 epoch were too overbaked. This checkpoint, at 1 epoch performed best. + + I'm pretty surprised how decent this came out since Qwen models aren't that great at writing, especially at this size. + overrides: + parameters: + model: lemon07r_VellumMini-0.1-Qwen3-14B-Q4_K_M.gguf + files: + - filename: lemon07r_VellumMini-0.1-Qwen3-14B-Q4_K_M.gguf + sha256: 7c56980b12c757e06bd4d4e99fca4eacf76fbad9bc46d59fde5fb62280157320 + uri: huggingface://bartowski/lemon07r_VellumMini-0.1-Qwen3-14B-GGUF/lemon07r_VellumMini-0.1-Qwen3-14B-Q4_K_M.gguf - &gemma3 url: "github:mudler/LocalAI/gallery/gemma.yaml@master" name: "gemma-3-27b-it"