From fa815f76c9ccc247fa3770e648ed42de52a420ff Mon Sep 17 00:00:00 2001 From: Concedo <39025047+LostRuins@users.noreply.github.com> Date: Sat, 2 Aug 2025 11:32:53 +0800 Subject: [PATCH] updated model recs (+1 squashed commits) Squashed commits: [3e0431ae1] updated model recs --- README.md | 4 ++-- klite.embd | 17 ++++++++++++++++- koboldcpp.py | 7 ++----- 3 files changed, 20 insertions(+), 8 deletions(-) diff --git a/README.md b/README.md index 9c73fba61..5e4ea3808 100644 --- a/README.md +++ b/README.md @@ -65,7 +65,7 @@ KoboldCpp can now also be run on Novita AI, a newer alternative GPU cloud provid ## Obtaining a GGUF model - KoboldCpp uses GGUF models. They are not included with KoboldCpp, but you can download GGUF files from other places such as [Bartowski's Huggingface](https://huggingface.co/bartowski). Search for "GGUF" on huggingface.co for plenty of compatible models in the `.gguf` format. -- For beginners, we recommend the models [Airoboros Mistral 7B](https://huggingface.co/TheBloke/airoboros-mistral2.2-7B-GGUF/resolve/main/airoboros-mistral2.2-7b.Q4_K_S.gguf) (smaller and weaker) or [Tiefighter 13B](https://huggingface.co/KoboldAI/LLaMA2-13B-Tiefighter-GGUF/resolve/main/LLaMA2-13B-Tiefighter.Q4_K_S.gguf) (larger model) or [Beepo 22B](https://huggingface.co/concedo/Beepo-22B-GGUF/resolve/main/Beepo-22B-Q4_K_S.gguf) (largest and most powerful) +- For beginners, we recommend the models [L3-8B-Stheno-v3.2](https://huggingface.co/bartowski/L3-8B-Stheno-v3.2-GGUF/resolve/main/L3-8B-Stheno-v3.2-Q4_K_S.gguf) (smaller and weaker) or [Tiefighter 13B](https://huggingface.co/KoboldAI/LLaMA2-13B-Tiefighter-GGUF/resolve/main/LLaMA2-13B-Tiefighter.Q4_K_S.gguf) (old but very versatile model) or [Gemma-3-27B Abliterated](https://huggingface.co/mlabonne/gemma-3-27b-it-abliterated-GGUF/resolve/main/gemma-3-27b-it-abliterated.q4_k_m.gguf) (largest and most powerful) - [Alternatively, you can download the tools to convert models to the GGUF format yourself here](https://kcpptools.concedo.workers.dev). Run `convert-hf-to-gguf.py` to convert them, then `quantize_gguf.exe` to quantize the result. - Other models for Whisper (speech recognition), Image Generation, Text to Speech or Image Recognition [can be found on the Wiki](https://github.com/LostRuins/koboldcpp/wiki#what-models-does-koboldcpp-support-what-architectures-are-supported) @@ -204,7 +204,7 @@ and it will install everything required. Alternatively, you can download the abo # Where can I download AI model files? - The best place to get GGUF text models is huggingface. For image models, CivitAI has a good selection. Here are some to get started. - - Text Generation: [Airoboros Mistral 7B](https://huggingface.co/TheBloke/airoboros-mistral2.2-7B-GGUF/resolve/main/airoboros-mistral2.2-7b.Q4_K_S.gguf) (smaller and weaker) or [Tiefighter 13B](https://huggingface.co/KoboldAI/LLaMA2-13B-Tiefighter-GGUF/resolve/main/LLaMA2-13B-Tiefighter.Q4_K_S.gguf) (larger model) or [Beepo 22B](https://huggingface.co/concedo/Beepo-22B-GGUF/resolve/main/Beepo-22B-Q4_K_S.gguf) (largest and most powerful) + - Text Generation: [L3-8B-Stheno-v3.2](https://huggingface.co/bartowski/L3-8B-Stheno-v3.2-GGUF/resolve/main/L3-8B-Stheno-v3.2-Q4_K_S.gguf) (smaller and weaker) or [Tiefighter 13B](https://huggingface.co/KoboldAI/LLaMA2-13B-Tiefighter-GGUF/resolve/main/LLaMA2-13B-Tiefighter.Q4_K_S.gguf) (old but very versatile model) or [Gemma-3-27B Abliterated](https://huggingface.co/mlabonne/gemma-3-27b-it-abliterated-GGUF/resolve/main/gemma-3-27b-it-abliterated.q4_k_m.gguf) (largest and most powerful) - Image Generation: [Anything v3](https://huggingface.co/admruul/anything-v3.0/resolve/main/Anything-V3.0-pruned-fp16.safetensors) or [Deliberate V2](https://huggingface.co/Yntec/Deliberate2/resolve/main/Deliberate_v2.safetensors) or [Dreamshaper SDXL](https://huggingface.co/Lykon/dreamshaper-xl-v2-turbo/resolve/main/DreamShaperXL_Turbo_v2_1.safetensors) - Image Recognition MMproj: [Pick the correct one for your model architecture here](https://huggingface.co/koboldcpp/mmproj/tree/main) - Speech Recognition: [Whisper models for Speech-To-Text](https://huggingface.co/koboldcpp/whisper/tree/main) diff --git a/klite.embd b/klite.embd index c1822fe99..00ecadc3d 100644 --- a/klite.embd +++ b/klite.embd @@ -13839,6 +13839,20 @@ Current version indicated by LITEVER below. document.getElementById("documentdb_snippetestimate").innerText = `Estimate of context usage: ${snippetEstimate / 3} tokens (${snippetEstimate} characters). Total chunks: ${numberOfChunks}. Cached chunks: ${Object.keys(embeddings_cache).length}`; } + function add_textdb_separator() + { + const textarea = document.getElementById('documentdb_data'); + const text = '[DOCUMENT BREAK]'; + const start = textarea.selectionStart; + const end = textarea.selectionEnd; + const before = textarea.value.substring(0, start); + const after = textarea.value.substring(end); + textarea.value = before + text + after; + const newPos = start + text.length; + textarea.selectionStart = textarea.selectionEnd = newPos; + textarea.focus(); // Keep focus + } + function confirm_memory() { current_memory = document.getElementById("memorytext").value; current_anote = document.getElementById("anotetext").value; @@ -24447,7 +24461,8 @@ Current version indicated by LITEVER below.
Please note, as with world info this can cause context reprocessing. It is recommended to have at least 8K context size and a good prompt processing speed to use this feature.

TextDB Storage?Paste as much raw text data here as you like. E.g. background information, reference documents, etc. This text will populate the database that will be chunked and searched by TextDB. This can be split using "[DOCUMENT BREAK]" to split the overall database into smaller documents, and titles can be added with "[DOCUMENT BREAK][Title of document]This is the content of the document".
+ class="helptext">Paste as much raw text data here as you like. E.g. background information, reference documents, etc. This text will populate the database that will be chunked and searched by TextDB. This can be split using "[DOCUMENT BREAK]" to split the overall database into smaller documents, and titles can be added with "[DOCUMENT BREAK][Title of document]This is the content of the document". +
diff --git a/koboldcpp.py b/koboldcpp.py index 5c3a1890d..72b64f97e 100644 --- a/koboldcpp.py +++ b/koboldcpp.py @@ -5769,9 +5769,6 @@ def show_gui(): import_vars(dict) pass - def display_help(): - LaunchWebbrowser("https://github.com/LostRuins/koboldcpp/wiki","Cannot launch help in browser.") - def display_help_models(): LaunchWebbrowser("https://github.com/LostRuins/koboldcpp/wiki#what-models-does-koboldcpp-support-what-architectures-are-supported","Cannot launch help in browser.") @@ -5783,7 +5780,7 @@ def show_gui(): ctk.CTkButton(tabs , text = "Update", fg_color="#9900cc", hover_color="#aa11dd", command = display_updates, width=90, height = 35 ).grid(row=1,column=0, stick="sw", padx= 5, pady=5) ctk.CTkButton(tabs , text = "Save Config", fg_color="#084a66", hover_color="#085a88", command = save_config_gui, width=60, height = 35 ).grid(row=1,column=1, stick="sw", padx= 5, pady=5) ctk.CTkButton(tabs , text = "Load Config", fg_color="#084a66", hover_color="#085a88", command = load_config_gui, width=60, height = 35 ).grid(row=1,column=1, stick="sw", padx= 92, pady=5) - ctk.CTkButton(tabs , text = "Help (Find Models)", fg_color="#992222", hover_color="#bb3333", command = display_help, width=100, height = 35 ).grid(row=1,column=1, stick="sw", padx= 180, pady=5) + ctk.CTkButton(tabs , text = "Help (Find Models)", fg_color="#992222", hover_color="#bb3333", command = display_help_models, width=100, height = 35 ).grid(row=1,column=1, stick="sw", padx= 180, pady=5) # start a thread that tries to get actual gpu names and layer counts gpuinfo_thread = threading.Thread(target=auto_set_backend_gui) @@ -5817,7 +5814,7 @@ def show_gui(): print("") time.sleep(0.5) if using_gui_launcher: - givehelp = show_gui_yesnobox("No Model Loaded","No text or image model file was selected. Cannot continue.\n\nDo you want help finding a GGUF model?") + givehelp = show_gui_yesnobox("No Model Loaded","No text or image model file was selected. Need a model to continue.\n\nDo you want help finding a GGUF model?") if givehelp == 'yes': display_help_models() else: