https://koboldcpp.dev/about/https://koboldcpp.dev/ai-horde/https://koboldcpp.dev/alternatives/https://koboldcpp.dev/amd/https://koboldcpp.dev/android/https://koboldcpp.dev/api/https://koboldcpp.dev/_astro/lab-lnx-api-swagger-01.-3DXP2ko_2aOOCq.pngKoboldCpp API Swagger page listing /v1/completions, /v1/chat/completions and other endpointshttps://koboldcpp.dev/_astro/lab-lnx-api-swagger-01.-3DXP2ko.pngKoboldCpp API Swagger page listing /v1/completions, /v1/chat/completions and other endpointshttps://koboldcpp.dev/api/kobold/https://koboldcpp.dev/api/openai/https://koboldcpp.dev/backends/https://koboldcpp.dev/build-from-source/https://koboldcpp.dev/chat-templates/https://koboldcpp.dev/cloud-gpu/https://koboldcpp.dev/colab/https://koboldcpp.dev/contact/https://koboldcpp.dev/context-size/https://koboldcpp.dev/corrections/https://koboldcpp.dev/docker/https://koboldcpp.dev/download/https://koboldcpp.dev/embeddings/https://koboldcpp.dev/_astro/gha-u24-lite-textdb-provider-01.DW4UNWLO_Z2m5AhJ.pngLite TextDB tab with the provider KoboldCpp Embeddings and a document text box holding two short documentshttps://koboldcpp.dev/_astro/gha-u24-lite-textdb-provider-01.DW4UNWLO.pngLite TextDB tab with the provider KoboldCpp Embeddings and a document text box holding two short documentshttps://koboldcpp.dev/_astro/gha-u24-lite-textdb-answer-01.CvgcgYmr_Z2b6650.pngLite chat where the question about the Falcon Hotel pastry chef is answered from the stored documentshttps://koboldcpp.dev/_astro/gha-u24-lite-textdb-answer-01.CvgcgYmr.pngLite chat where the question about the Falcon Hotel pastry chef is answered from the stored documentshttps://koboldcpp.dev/_astro/gha-u24-lite-context-textdb-01.CZ-8f78n_wU1kb.pngLite Context Data dialog TextDB tab with a provider dropdown set to Disabled, search results 3 and snippet size 800https://koboldcpp.dev/_astro/gha-u24-lite-context-textdb-01.CZ-8f78n.pngLite Context Data dialog TextDB tab with a provider dropdown set to Disabled, search results 3 and snippet size 800https://koboldcpp.dev/_astro/gha-u24-lite-settings-tools-01.BUH2ZXU7_ZBbkyi.pngLite Settings Tools tab with MCP tool calling, MCP server URLs and the Enable Websearch optionhttps://koboldcpp.dev/_astro/gha-u24-lite-settings-tools-01.BUH2ZXU7.pngLite Settings Tools tab with MCP tool calling, MCP server URLs and the Enable Websearch optionhttps://koboldcpp.dev/errors/https://koboldcpp.dev/faq/https://koboldcpp.dev/flags/https://koboldcpp.dev/getting-started/https://koboldcpp.dev/_astro/lab-w25-launcher-quick-01.B1HBxCji_Z1xxqmd.pngKoboldCpp launcher Quick Launch tab with Use CPU backend, ContextShift and FlashAttention ticked, context 16384https://koboldcpp.dev/_astro/lab-w25-launcher-quick-01.B1HBxCji.pngKoboldCpp launcher Quick Launch tab with Use CPU backend, ContextShift and FlashAttention ticked, context 16384https://koboldcpp.dev/_astro/lab-lnx-lite-welcome-01.2vYMfMxM_2bu4Yt.pngWelcome To KoboldAI Lite dialog offering Classic, Aesthetic and Corpo UI styleshttps://koboldcpp.dev/_astro/lab-lnx-lite-welcome-01.2vYMfMxM.pngWelcome To KoboldAI Lite dialog offering Classic, Aesthetic and Corpo UI styleshttps://koboldcpp.dev/_astro/lab-lnx-lite-firstprompt-01.gfO6vge3_4Cmgt.pngKoboldAI Lite with the welcome dialog open and an unsent prompt in the input boxhttps://koboldcpp.dev/_astro/lab-lnx-lite-firstprompt-01.gfO6vge3.pngKoboldAI Lite with the welcome dialog open and an unsent prompt in the input boxhttps://koboldcpp.dev/gpu-layers/https://koboldcpp.dev/_astro/lab-w25-launcher-quick-01.B1HBxCji_Z1xxqmd.pngKoboldCpp launcher Quick Launch tab with Use CPU backend, ContextShift and FlashAttention ticked, context 16384https://koboldcpp.dev/_astro/lab-w25-launcher-quick-01.B1HBxCji.pngKoboldCpp launcher Quick Launch tab with Use CPU backend, ContextShift and FlashAttention ticked, context 16384https://koboldcpp.dev/_astro/gha-w25-launcher-hardware-01.BDX6DZ2h_Z16TSiL.pngKoboldCpp launcher, Hardware tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-hardware-01.BDX6DZ2h.pngKoboldCpp launcher, Hardware tab, default settingshttps://koboldcpp.dev/how-we-test/https://koboldcpp.dev/huggingface/https://koboldcpp.dev/image-generation/https://koboldcpp.dev/_astro/gha-u24-lite-addfile-menu-01.ERaWJd4e_Zbmg15.pngLite Add Audio or Image dialog with Generate Image, Generate Speech, Capture From, Upload a File and Customize Media Settingshttps://koboldcpp.dev/_astro/gha-u24-lite-addfile-menu-01.ERaWJd4e.pngLite Add Audio or Image dialog with Generate Image, Generate Speech, Capture From, Upload a File and Customize Media Settingshttps://koboldcpp.dev/_astro/gha-u24-lite-image-generated-01.YePJJXcG_1JpHzS.pngLite chat with a model reply and a small generated picture of a woman reading a book under ithttps://koboldcpp.dev/_astro/gha-u24-lite-image-generated-01.YePJJXcG.pngLite chat with a model reply and a small generated picture of a woman reading a book under ithttps://koboldcpp.dev/_astro/gha-u24-lite-settings-media-01.rLn2TAuM_1eYHn0.pngLite Settings Media tab with the KCPP, Forge, A1111 image backend, the sdxs model, step count, CFG scale and text to speechhttps://koboldcpp.dev/_astro/gha-u24-lite-settings-media-01.rLn2TAuM.pngLite Settings Media tab with the KCPP, Forge, A1111 image backend, the sdxs model, step count, CFG scale and text to speechhttps://koboldcpp.dev/_astro/gha-u24-stableui-01.Don0v0PF_CcOLa.pngStableUI page with Generate, Images, About and Options tabs, a prompt box, seed, sampler and schedulerhttps://koboldcpp.dev/_astro/gha-u24-stableui-01.Don0v0PF.pngStableUI page with Generate, Images, About and Options tabs, a prompt box, seed, sampler and schedulerhttps://koboldcpp.dev/_astro/gha-u24-musicui-01.DZSTr5Ap_ZNBSy7.pngMusicUI page of KoboldCpphttps://koboldcpp.dev/_astro/gha-u24-musicui-01.DZSTr5Ap.pngMusicUI page of KoboldCpphttps://koboldcpp.dev/https://koboldcpp.dev/integrations/https://koboldcpp.dev/is-koboldcpp-safe/https://koboldcpp.dev/kcpps/https://koboldcpp.dev/kobold-lite/https://koboldcpp.dev/_astro/lab-lnx-lite-welcome-01.2vYMfMxM_2bu4Yt.pngWelcome To KoboldAI Lite dialog offering Classic, Aesthetic and Corpo UI styleshttps://koboldcpp.dev/_astro/lab-lnx-lite-welcome-01.2vYMfMxM.pngWelcome To KoboldAI Lite dialog offering Classic, Aesthetic and Corpo UI styleshttps://koboldcpp.dev/_astro/gha-u24-lite-settings-general-01.BYKYRoBj_ZP79Gd.pngKoboldAI Lite Settings dialog, General tab, with usage mode Instruct Mode and the KoboldCppAutomatic instruct tag presethttps://koboldcpp.dev/_astro/gha-u24-lite-settings-general-01.BYKYRoBj.pngKoboldAI Lite Settings dialog, General tab, with usage mode Instruct Mode and the KoboldCppAutomatic instruct tag presethttps://koboldcpp.dev/_astro/gha-u24-lite-settings-tokens-01.CpWSKOJU_Z1XuvwH.pngLite Settings Tokens tab with collapsed sections for stop sequences, logit biases and thinking tagshttps://koboldcpp.dev/_astro/gha-u24-lite-settings-tokens-01.CpWSKOJU.pngLite Settings Tokens tab with collapsed sections for stop sequences, logit biases and thinking tagshttps://koboldcpp.dev/_astro/gha-u24-lite-settings-misc-01.geYsug4q_Z1ug7OJ.pngLite Settings Misc tab with autosave, EOS token ban, token limit multiplier and reset all settingshttps://koboldcpp.dev/_astro/gha-u24-lite-settings-misc-01.geYsug4q.pngLite Settings Misc tab with autosave, EOS token ban, token limit multiplier and reset all settingshttps://koboldcpp.dev/_astro/lab-lnx-lite-mobile-01.CxgOhiXc_Z1yffM0.pngKoboldAI Lite welcome screen on a narrow mobile viewporthttps://koboldcpp.dev/_astro/lab-lnx-lite-mobile-01.CxgOhiXc.pngKoboldAI Lite welcome screen on a narrow mobile viewporthttps://koboldcpp.dev/_astro/gha-u24-lite-classic-mobile-01.BzdtDPTc_QoVRU.pngKoboldAI Lite chat screen on a 390 pixel wide phone screenhttps://koboldcpp.dev/_astro/gha-u24-lite-classic-mobile-01.BzdtDPTc.pngKoboldAI Lite chat screen on a 390 pixel wide phone screenhttps://koboldcpp.dev/koboldai/https://koboldcpp.dev/_astro/lab-lnx-lite-welcome-01.2vYMfMxM_2bu4Yt.pngWelcome To KoboldAI Lite dialog offering Classic, Aesthetic and Corpo UI styleshttps://koboldcpp.dev/_astro/lab-lnx-lite-welcome-01.2vYMfMxM.pngWelcome To KoboldAI Lite dialog offering Classic, Aesthetic and Corpo UI styleshttps://koboldcpp.dev/launcher/https://koboldcpp.dev/_astro/gha-w25-launcher-quick-01.CNheZ9Lp_24A96n.webpKoboldCpp launcher, Quick Launch tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-quick-01.CNheZ9Lp_Z19tC17.pngKoboldCpp launcher, Quick Launch tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-quick-01.CNheZ9Lp.pngKoboldCpp launcher, Quick Launch tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-hardware-01.BDX6DZ2h_Z16TSiL.pngKoboldCpp launcher, Hardware tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-hardware-01.BDX6DZ2h.pngKoboldCpp launcher, Hardware tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-context-01.BK9tFZSi_1jIXKR.pngKoboldCpp launcher, Context tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-context-01.BK9tFZSi.pngKoboldCpp launcher, Context tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-files-01.SV9EYA12_Z1L7ycH.pngKoboldCpp launcher, Loaded Files tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-files-01.SV9EYA12.pngKoboldCpp launcher, Loaded Files tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-network-01.DWUIz2n8_ZyOm5t.pngKoboldCpp launcher, Network tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-network-01.DWUIz2n8.pngKoboldCpp launcher, Network tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-horde-01.DGglxVyB_Z1mbChT.pngKoboldCpp launcher, Horde Worker tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-horde-01.DGglxVyB.pngKoboldCpp launcher, Horde Worker tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-imagegen-01.C0XnZP7I_ZKScne.pngKoboldCpp launcher, Image Gen tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-imagegen-01.C0XnZP7I.pngKoboldCpp launcher, Image Gen tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-audio-01.DNmFjz-t_Z1HL2Xl.pngKoboldCpp launcher, Audio tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-audio-01.DNmFjz-t.pngKoboldCpp launcher, Audio tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-agent-01.S0VxJFfq_ZP8zFp.pngKoboldCpp launcher, Agent tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-agent-01.S0VxJFfq.pngKoboldCpp launcher, Agent tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-admin-01.C5T4yam-_1dtl06.pngKoboldCpp launcher, Admin tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-admin-01.C5T4yam-.pngKoboldCpp launcher, Admin tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-extra-01.DQ90xciz_EraqY.pngKoboldCpp launcher, Extra tab, default settingshttps://koboldcpp.dev/_astro/gha-w25-launcher-extra-01.DQ90xciz.pngKoboldCpp launcher, Extra tab, default settingshttps://koboldcpp.dev/linux/https://koboldcpp.dev/mac/https://koboldcpp.dev/model-wont-load/https://koboldcpp.dev/models/https://koboldcpp.dev/multi-gpu/https://koboldcpp.dev/not-using-gpu/https://koboldcpp.dev/_astro/lab-w25-launcher-quick-01.B1HBxCji_Z1xxqmd.pngKoboldCpp launcher Quick Launch tab with Use CPU backend, ContextShift and FlashAttention ticked, context 16384https://koboldcpp.dev/_astro/lab-w25-launcher-quick-01.B1HBxCji.pngKoboldCpp launcher Quick Launch tab with Use CPU backend, ContextShift and FlashAttention ticked, context 16384https://koboldcpp.dev/nvidia/https://koboldcpp.dev/out-of-memory/https://koboldcpp.dev/privacy/https://koboldcpp.dev/quantization/https://koboldcpp.dev/releases/https://koboldcpp.dev/remote-access/https://koboldcpp.dev/requirements/https://koboldcpp.dev/samplers/https://koboldcpp.dev/_astro/gha-u24-lite-settings-samplers-01.BPO_isRi_Zh7vwd.pngLite Settings Samplers tab: context 16384, max output 2048, temperature 0.75, repeat penalty 1.05, top-p 0.92, top-k 100https://koboldcpp.dev/_astro/gha-u24-lite-settings-samplers-01.BPO_isRi.pngLite Settings Samplers tab: context 16384, max output 2048, temperature 0.75, repeat penalty 1.05, top-p 0.92, top-k 100https://koboldcpp.dev/sillytavern/https://koboldcpp.dev/_astro/gha-st-firstrun-01.BHmG0wwn_ZUgYu6.pngSillyTavern welcome dialog asking for a persona namehttps://koboldcpp.dev/_astro/gha-st-firstrun-01.BHmG0wwn.pngSillyTavern welcome dialog asking for a persona namehttps://koboldcpp.dev/_astro/gha-st-connect-01.OZTqC3uP_ZW3pxh.pngSillyTavern API Connections panelhttps://koboldcpp.dev/_astro/gha-st-connect-01.OZTqC3uP.pngSillyTavern API Connections panelhttps://koboldcpp.dev/_astro/gha-st-connect-02.D_IAPM6o_Z1q3c6t.pngSillyTavern Text Completion settings with KoboldCpp selectedhttps://koboldcpp.dev/_astro/gha-st-connect-02.D_IAPM6o.pngSillyTavern Text Completion settings with KoboldCpp selectedhttps://koboldcpp.dev/_astro/gha-st-textcomp-correct-01.C7ZlUdb8_Zg7Vfk.pngSillyTavern connected to KoboldCpp showing the loaded model namehttps://koboldcpp.dev/_astro/gha-st-textcomp-correct-01.C7ZlUdb8.pngSillyTavern connected to KoboldCpp showing the loaded model namehttps://koboldcpp.dev/_astro/gha-st-chatcomp-correct-01.D-aCz0RE_ZhofKV.pngSillyTavern Chat Completion custom endpoint with /v1https://koboldcpp.dev/_astro/gha-st-chatcomp-correct-01.D-aCz0RE.pngSillyTavern Chat Completion custom endpoint with /v1https://koboldcpp.dev/_astro/gha-st-chatcomp-missing-v1-01.BvroQ6OM_14Hi1y.pngSillyTavern Chat Completion custom endpoint validated against KoboldCpphttps://koboldcpp.dev/_astro/gha-st-chatcomp-missing-v1-01.BvroQ6OM.pngSillyTavern Chat Completion custom endpoint validated against KoboldCpphttps://koboldcpp.dev/_astro/gha-st-textcomp-v1-suffix-01.Ch6JVU4l_Zv0e2i.pngSillyTavern connected with a /v1 URL and a Missing key markerhttps://koboldcpp.dev/_astro/gha-st-textcomp-v1-suffix-01.Ch6JVU4l.pngSillyTavern connected with a /v1 URL and a Missing key markerhttps://koboldcpp.dev/_astro/gha-st-chat-01.B08wWW-p_Zb6xhM.pngSillyTavern chat with a reply generated by KoboldCpphttps://koboldcpp.dev/_astro/gha-st-chat-01.B08wWW-p.pngSillyTavern chat with a reply generated by KoboldCpphttps://koboldcpp.dev/speed/https://koboldcpp.dev/terms/https://koboldcpp.dev/tools/api-tester/https://koboldcpp.dev/tools/config-builder/https://koboldcpp.dev/tools/https://koboldcpp.dev/tools/log-analyzer/https://koboldcpp.dev/tools/vram-calculator/https://koboldcpp.dev/troubleshooting/https://koboldcpp.dev/vision/https://koboldcpp.dev/_astro/gha-u24-lite-upload-image-01.BP7kvgzk_TStX6.pngLite chat showing an uploaded picture of a red circle and a blue rectangle in the user turnhttps://koboldcpp.dev/_astro/gha-u24-lite-upload-image-01.BP7kvgzk.pngLite chat showing an uploaded picture of a red circle and a blue rectangle in the user turnhttps://koboldcpp.dev/_astro/gha-u24-lite-vision-reply-01.CjYdRXUL_3m6Fi.pngLite chat with the uploaded red circle and blue rectangle and the tiny vision model answering Triangleshttps://koboldcpp.dev/_astro/gha-u24-lite-vision-reply-01.CjYdRXUL.pngLite chat with the uploaded red circle and blue rectangle and the tiny vision model answering Triangleshttps://koboldcpp.dev/voice/https://koboldcpp.dev/_astro/gha-u24-lite-settings-media-01.rLn2TAuM_1eYHn0.pngLite Settings Media tab with the KCPP, Forge, A1111 image backend, the sdxs model, step count, CFG scale and text to speechhttps://koboldcpp.dev/_astro/gha-u24-lite-settings-media-01.rLn2TAuM.pngLite Settings Media tab with the KCPP, Forge, A1111 image backend, the sdxs model, step count, CFG scale and text to speechhttps://koboldcpp.dev/_astro/gha-u24-lite-stt-result-01.BD1L0zuR_tJPac.pngLite chat with spoken phrases transcribed into user turns, each followed by a model reply, and a microphone button lithttps://koboldcpp.dev/_astro/gha-u24-lite-stt-result-01.BD1L0zuR.pngLite chat with spoken phrases transcribed into user turns, each followed by a model reply, and a microphone button lithttps://koboldcpp.dev/_astro/gha-u24-lite-tts-settings-01.B4h6lciD_19qC0R.pngLite Media settings with Text to speech set to KoboldCpp TTS API, with voice choices and a Test buttonhttps://koboldcpp.dev/_astro/gha-u24-lite-tts-settings-01.B4h6lciD.pngLite Media settings with Text to speech set to KoboldCpp TTS API, with voice choices and a Test buttonhttps://koboldcpp.dev/vs/llama-cpp/https://koboldcpp.dev/vs/lm-studio/https://koboldcpp.dev/vs/ollama/https://koboldcpp.dev/vs/text-generation-webui/https://koboldcpp.dev/what-is-koboldcpp/https://koboldcpp.dev/_astro/lab-lnx-lite-welcome-01.2vYMfMxM_2bu4Yt.pngWelcome To KoboldAI Lite dialog offering Classic, Aesthetic and Corpo UI styleshttps://koboldcpp.dev/_astro/lab-lnx-lite-welcome-01.2vYMfMxM.pngWelcome To KoboldAI Lite dialog offering Classic, Aesthetic and Corpo UI styleshttps://koboldcpp.dev/windows/https://koboldcpp.dev/_astro/lab-w25-launcher-quick-01.B1HBxCji_Z1xxqmd.pngKoboldCpp launcher Quick Launch tab with Use CPU backend, ContextShift and FlashAttention ticked, context 16384https://koboldcpp.dev/_astro/lab-w25-launcher-quick-01.B1HBxCji.pngKoboldCpp launcher Quick Launch tab with Use CPU backend, ContextShift and FlashAttention ticked, context 16384https://koboldcpp.dev/wont-start/