Merge branch 'master' of github.com:Saavrm26/LocalAI into resume_download

Makefile

@@ -8,7 +8,7 @@ DETECT_LIBS?=true
 # llama.cpp versions
 GOLLAMA_REPO?=https://github.com/go-skynet/go-llama.cpp
 GOLLAMA_VERSION?=2b57a8ae43e4699d3dc5d1496a1ccd42922993be
-CPPLLAMA_VERSION?=9394bbd484f802ce80d2858033583af3ef700d25
+CPPLLAMA_VERSION?=53ff6b9b9fb25ed0ec0a213e05534fe7c3d0040f
 
 # whisper.cpp version
 WHISPER_REPO?=https://github.com/ggerganov/whisper.cpp

core/http/app.go

@@ -87,6 +87,8 @@ func API(application *application.Application) (*fiber.App, error) {
 
 	router := fiber.New(fiberCfg)
 
+	router.Use(middleware.StripPathPrefix())
+
 	router.Hooks().OnListen(func(listenData fiber.ListenData) error {
 		scheme := "http"
 		if listenData.TLS {

core/http/app_test.go

@@ -237,6 +237,31 @@ func postInvalidRequest(url string) (error, int) {
 	return nil, resp.StatusCode
 }
 
+func getRequest(url string, header http.Header) (error, int, []byte) {
+
+	req, err := http.NewRequest("GET", url, nil)
+	if err != nil {
+		return err, -1, nil
+	}
+
+	req.Header = header
+
+	client := &http.Client{}
+	resp, err := client.Do(req)
+	if err != nil {
+		return err, -1, nil
+	}
+
+	defer resp.Body.Close()
+
+	body, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return err, -1, nil
+	}
+
+	return nil, resp.StatusCode, body
+}
+
 const bertEmbeddingsURL = `https://gist.githubusercontent.com/mudler/0a080b166b87640e8644b09c2aee6e3b/raw/f0e8c26bb72edc16d9fbafbfd6638072126ff225/bert-embeddings-gallery.yaml`
 
 //go:embed backend-assets/*
@@ -345,6 +370,33 @@ var _ = Describe("API test", func() {
 			})
 		})
 
+		Context("URL routing Tests", func() {
+			It("Should support reverse-proxy when unauthenticated", func() {
+
+				err, sc, body := getRequest("http://127.0.0.1:9090/myprefix/", http.Header{
+					"X-Forwarded-Proto":  {"https"},
+					"X-Forwarded-Host":   {"example.org"},
+					"X-Forwarded-Prefix": {"/myprefix/"},
+				})
+				Expect(err).To(BeNil(), "error")
+				Expect(sc).To(Equal(401), "status code")
+				Expect(string(body)).To(ContainSubstring(`<base href="https://example.org/myprefix/" />`), "body")
+			})
+
+			It("Should support reverse-proxy when authenticated", func() {
+
+				err, sc, body := getRequest("http://127.0.0.1:9090/myprefix/", http.Header{
+					"Authorization":      {bearerKey},
+					"X-Forwarded-Proto":  {"https"},
+					"X-Forwarded-Host":   {"example.org"},
+					"X-Forwarded-Prefix": {"/myprefix/"},
+				})
+				Expect(err).To(BeNil(), "error")
+				Expect(sc).To(Equal(200), "status code")
+				Expect(string(body)).To(ContainSubstring(`<base href="https://example.org/myprefix/" />`), "body")
+			})
+		})
+
 		Context("Applying models", func() {
 
 			It("applies models from a gallery", func() {

core/http/elements/buttons.go

@@ -16,7 +16,7 @@ func installButton(galleryName string) elem.Node {
 			"class":                 "float-right inline-block rounded bg-primary px-6 pb-2.5 mb-3 pt-2.5 text-xs font-medium uppercase leading-normal text-white shadow-primary-3 transition duration-150 ease-in-out hover:bg-primary-accent-300 hover:shadow-primary-2 focus:bg-primary-accent-300 focus:shadow-primary-2 focus:outline-none focus:ring-0 active:bg-primary-600 active:shadow-primary-2 dark:shadow-black/30 dark:hover:shadow-dark-strong dark:focus:shadow-dark-strong dark:active:shadow-dark-strong",
 			"hx-swap":               "outerHTML",
 			// post the Model ID as param
-			"hx-post": "/browse/install/model/" + galleryName,
+			"hx-post": "browse/install/model/" + galleryName,
 		},
 		elem.I(
 			attrs.Props{
@@ -36,7 +36,7 @@ func reInstallButton(galleryName string) elem.Node {
 			"hx-target":             "#action-div-" + dropBadChars(galleryName),
 			"hx-swap":               "outerHTML",
 			// post the Model ID as param
-			"hx-post": "/browse/install/model/" + galleryName,
+			"hx-post": "browse/install/model/" + galleryName,
 		},
 		elem.I(
 			attrs.Props{
@@ -80,7 +80,7 @@ func deleteButton(galleryID string) elem.Node {
 			"hx-target":             "#action-div-" + dropBadChars(galleryID),
 			"hx-swap":               "outerHTML",
 			// post the Model ID as param
-			"hx-post": "/browse/delete/model/" + galleryID,
+			"hx-post": "browse/delete/model/" + galleryID,
 		},
 		elem.I(
 			attrs.Props{

core/http/elements/gallery.go

@@ -47,7 +47,7 @@ func searchableElement(text, icon string) elem.Node {
 					//	"value":     text,
 					//"class":     "inline-block bg-gray-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2",
 					"href":      "#!",
-					"hx-post":   "/browse/search/models",
+					"hx-post":   "browse/search/models",
 					"hx-target": "#search-results",
 					// TODO: this doesn't work
 					//	"hx-vals":      `{ \"search\": \"` + text + `\" }`,

core/http/elements/progressbar.go

@@ -64,7 +64,7 @@ func StartProgressBar(uid, progress, text string) string {
 	return elem.Div(
 		attrs.Props{
 			"hx-trigger": "done",
-			"hx-get":     "/browse/job/" + uid,
+			"hx-get":     "browse/job/" + uid,
 			"hx-swap":    "outerHTML",
 			"hx-target":  "this",
 		},
@@ -77,7 +77,7 @@ func StartProgressBar(uid, progress, text string) string {
 			},
 			elem.Text(bluemonday.StrictPolicy().Sanitize(text)), //Perhaps overly defensive
 			elem.Div(attrs.Props{
-				"hx-get":     "/browse/job/progress/" + uid,
+				"hx-get":     "browse/job/progress/" + uid,
 				"hx-trigger": "every 600ms",
 				"hx-target":  "this",
 				"hx-swap":    "innerHTML",

core/http/endpoints/explorer/dashboard.go

@@ -6,6 +6,7 @@ import (
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/explorer"
+	"github.com/mudler/LocalAI/core/http/utils"
 	"github.com/mudler/LocalAI/internal"
 )
 
@@ -14,6 +15,7 @@ func Dashboard() func(*fiber.Ctx) error {
 		summary := fiber.Map{
 			"Title":   "LocalAI API - " + internal.PrintableVersion(),
 			"Version": internal.PrintableVersion(),
+			"BaseURL": utils.BaseURL(c),
 		}
 
 		if string(c.Context().Request.Header.ContentType()) == "application/json" || len(c.Accepts("html")) == 0 {

core/http/endpoints/localai/gallery.go

@@ -9,6 +9,7 @@ import (
 	"github.com/google/uuid"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/gallery"
+	"github.com/mudler/LocalAI/core/http/utils"
 	"github.com/mudler/LocalAI/core/schema"
 	"github.com/mudler/LocalAI/core/services"
 	"github.com/rs/zerolog/log"
@@ -82,7 +83,8 @@ func (mgs *ModelGalleryEndpointService) ApplyModelGalleryEndpoint() func(c *fibe
 			Galleries:        mgs.galleries,
 			ConfigURL:        input.ConfigURL,
 		}
-		return c.JSON(schema.GalleryResponse{ID: uuid.String(), StatusURL: c.BaseURL() + "/models/jobs/" + uuid.String()})
+
+		return c.JSON(schema.GalleryResponse{ID: uuid.String(), StatusURL: fmt.Sprintf("%smodels/jobs/%s", utils.BaseURL(c), uuid.String())})
 	}
 }
 
@@ -105,7 +107,7 @@ func (mgs *ModelGalleryEndpointService) DeleteModelGalleryEndpoint() func(c *fib
 			return err
 		}
 
-		return c.JSON(schema.GalleryResponse{ID: uuid.String(), StatusURL: c.BaseURL() + "/models/jobs/" + uuid.String()})
+		return c.JSON(schema.GalleryResponse{ID: uuid.String(), StatusURL: fmt.Sprintf("%smodels/jobs/%s", utils.BaseURL(c), uuid.String())})
 	}
 }
 

core/http/endpoints/localai/welcome.go

@@ -4,6 +4,7 @@ import (
 	"github.com/gofiber/fiber/v2"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/gallery"
+	"github.com/mudler/LocalAI/core/http/utils"
 	"github.com/mudler/LocalAI/core/p2p"
 	"github.com/mudler/LocalAI/core/services"
 	"github.com/mudler/LocalAI/internal"
@@ -32,6 +33,7 @@ func WelcomeEndpoint(appConfig *config.ApplicationConfig,
 		summary := fiber.Map{
 			"Title":             "LocalAI API - " + internal.PrintableVersion(),
 			"Version":           internal.PrintableVersion(),
+			"BaseURL":           utils.BaseURL(c),
 			"Models":            modelsWithoutConfig,
 			"ModelsConfig":      backendConfigs,
 			"GalleryConfig":     galleryConfigs,

core/http/explorer.go

@@ -7,6 +7,7 @@ import (
 	"github.com/gofiber/fiber/v2/middleware/favicon"
 	"github.com/gofiber/fiber/v2/middleware/filesystem"
 	"github.com/mudler/LocalAI/core/explorer"
+	"github.com/mudler/LocalAI/core/http/middleware"
 	"github.com/mudler/LocalAI/core/http/routes"
 )
 
@@ -22,6 +23,7 @@ func Explorer(db *explorer.Database) *fiber.App {
 
 	app := fiber.New(fiberCfg)
 
+	app.Use(middleware.StripPathPrefix())
 	routes.RegisterExplorerRoutes(app, db)
 
 	httpFS := http.FS(embedDirStatic)

core/http/middleware/auth.go

@@ -8,6 +8,7 @@ import (
 	"github.com/gofiber/fiber/v2"
 	"github.com/gofiber/fiber/v2/middleware/keyauth"
 	"github.com/mudler/LocalAI/core/config"
+	"github.com/mudler/LocalAI/core/http/utils"
 )
 
 // This file contains the configuration generators and handler functions that are used along with the fiber/keyauth middleware
@@ -39,7 +40,9 @@ func getApiKeyErrorHandler(applicationConfig *config.ApplicationConfig) fiber.Er
 			if applicationConfig.OpaqueErrors {
 				return ctx.SendStatus(401)
 			}
-			return ctx.Status(401).Render("views/login", nil)
+			return ctx.Status(401).Render("views/login", fiber.Map{
+				"BaseURL": utils.BaseURL(ctx),
+			})
 		}
 		if applicationConfig.OpaqueErrors {
 			return ctx.SendStatus(500)

core/http/middleware/strippathprefix.go

@@ -0,0 +1,36 @@
+package middleware
+
+import (
+	"strings"
+
+	"github.com/gofiber/fiber/v2"
+)
+
+// StripPathPrefix returns a middleware that strips a path prefix from the request path.
+// The path prefix is obtained from the X-Forwarded-Prefix HTTP request header.
+func StripPathPrefix() fiber.Handler {
+	return func(c *fiber.Ctx) error {
+		for _, prefix := range c.GetReqHeaders()["X-Forwarded-Prefix"] {
+			if prefix != "" {
+				path := c.Path()
+				pos := len(prefix)
+
+				if prefix[pos-1] == '/' {
+					pos--
+				} else {
+					prefix += "/"
+				}
+
+				if strings.HasPrefix(path, prefix) {
+					c.Path(path[pos:])
+					break
+				} else if prefix[:pos] == path {
+					c.Redirect(prefix)
+					return nil
+				}
+			}
+		}
+
+		return c.Next()
+	}
+}

core/http/middleware/strippathprefix_test.go

@@ -0,0 +1,121 @@
+package middleware
+
+import (
+	"net/http/httptest"
+	"testing"
+
+	"github.com/gofiber/fiber/v2"
+	"github.com/stretchr/testify/require"
+)
+
+func TestStripPathPrefix(t *testing.T) {
+	var actualPath string
+
+	app := fiber.New()
+
+	app.Use(StripPathPrefix())
+
+	app.Get("/hello/world", func(c *fiber.Ctx) error {
+		actualPath = c.Path()
+		return nil
+	})
+
+	app.Get("/", func(c *fiber.Ctx) error {
+		actualPath = c.Path()
+		return nil
+	})
+
+	for _, tc := range []struct {
+		name         string
+		path         string
+		prefixHeader []string
+		expectStatus int
+		expectPath   string
+	}{
+		{
+			name:         "without prefix and header",
+			path:         "/hello/world",
+			expectStatus: 200,
+			expectPath:   "/hello/world",
+		},
+		{
+			name:         "without prefix and headers on root path",
+			path:         "/",
+			expectStatus: 200,
+			expectPath:   "/",
+		},
+		{
+			name:         "without prefix but header",
+			path:         "/hello/world",
+			prefixHeader: []string{"/otherprefix/"},
+			expectStatus: 200,
+			expectPath:   "/hello/world",
+		},
+		{
+			name:         "with prefix but non-matching header",
+			path:         "/prefix/hello/world",
+			prefixHeader: []string{"/otherprefix/"},
+			expectStatus: 404,
+		},
+		{
+			name:         "with prefix and matching header",
+			path:         "/myprefix/hello/world",
+			prefixHeader: []string{"/myprefix/"},
+			expectStatus: 200,
+			expectPath:   "/hello/world",
+		},
+		{
+			name:         "with prefix and 1st header matching",
+			path:         "/myprefix/hello/world",
+			prefixHeader: []string{"/myprefix/", "/otherprefix/"},
+			expectStatus: 200,
+			expectPath:   "/hello/world",
+		},
+		{
+			name:         "with prefix and 2nd header matching",
+			path:         "/myprefix/hello/world",
+			prefixHeader: []string{"/otherprefix/", "/myprefix/"},
+			expectStatus: 200,
+			expectPath:   "/hello/world",
+		},
+		{
+			name:         "with prefix and header not ending with slash",
+			path:         "/myprefix/hello/world",
+			prefixHeader: []string{"/myprefix"},
+			expectStatus: 200,
+			expectPath:   "/hello/world",
+		},
+		{
+			name:         "with prefix and non-matching header not ending with slash",
+			path:         "/myprefix-suffix/hello/world",
+			prefixHeader: []string{"/myprefix"},
+			expectStatus: 404,
+		},
+		{
+			name:         "redirect when prefix does not end with a slash",
+			path:         "/myprefix",
+			prefixHeader: []string{"/myprefix"},
+			expectStatus: 302,
+			expectPath:   "/myprefix/",
+		},
+	} {
+		t.Run(tc.name, func(t *testing.T) {
+			actualPath = ""
+			req := httptest.NewRequest("GET", tc.path, nil)
+			if tc.prefixHeader != nil {
+				req.Header["X-Forwarded-Prefix"] = tc.prefixHeader
+			}
+
+			resp, err := app.Test(req, -1)
+
+			require.NoError(t, err)
+			require.Equal(t, tc.expectStatus, resp.StatusCode, "response status code")
+
+			if tc.expectStatus == 200 {
+				require.Equal(t, tc.expectPath, actualPath, "rewritten path")
+			} else if tc.expectStatus == 302 {
+				require.Equal(t, tc.expectPath, resp.Header.Get("Location"), "redirect location")
+			}
+		})
+	}
+}

core/http/render.go

@@ -10,6 +10,7 @@ import (
 	"github.com/gofiber/fiber/v2"
 	fiberhtml "github.com/gofiber/template/html/v2"
 	"github.com/microcosm-cc/bluemonday"
+	"github.com/mudler/LocalAI/core/http/utils"
 	"github.com/mudler/LocalAI/core/schema"
 	"github.com/russross/blackfriday"
 )
@@ -26,7 +27,9 @@ func notFoundHandler(c *fiber.Ctx) error {
 		})
 	} else {
 		// The client expects an HTML response
-		return c.Status(fiber.StatusNotFound).Render("views/404", fiber.Map{})
+		return c.Status(fiber.StatusNotFound).Render("views/404", fiber.Map{
+			"BaseURL": utils.BaseURL(c),
+		})
 	}
 }
 

core/http/routes/ui.go

@@ -6,20 +6,21 @@ import (
 	"sort"
 	"strings"
 
-	"github.com/microcosm-cc/bluemonday"
 	"github.com/mudler/LocalAI/core/config"
 	"github.com/mudler/LocalAI/core/gallery"
 	"github.com/mudler/LocalAI/core/http/elements"
 	"github.com/mudler/LocalAI/core/http/endpoints/localai"
+	"github.com/mudler/LocalAI/core/http/utils"
 	"github.com/mudler/LocalAI/core/p2p"
 	"github.com/mudler/LocalAI/core/services"
 	"github.com/mudler/LocalAI/internal"
 	"github.com/mudler/LocalAI/pkg/model"
 	"github.com/mudler/LocalAI/pkg/xsync"
-	"github.com/rs/zerolog/log"
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/google/uuid"
+	"github.com/microcosm-cc/bluemonday"
+	"github.com/rs/zerolog/log"
 )
 
 type modelOpCache struct {
@@ -91,6 +92,7 @@ func RegisterUIRoutes(app *fiber.App,
 		app.Get("/p2p", func(c *fiber.Ctx) error {
 			summary := fiber.Map{
 				"Title":   "LocalAI - P2P dashboard",
+				"BaseURL": utils.BaseURL(c),
 				"Version": internal.PrintableVersion(),
 				//"Nodes":          p2p.GetAvailableNodes(""),
 				//"FederatedNodes": p2p.GetAvailableNodes(p2p.FederatedID),
@@ -149,6 +151,7 @@ func RegisterUIRoutes(app *fiber.App,
 
 			summary := fiber.Map{
 				"Title":            "LocalAI - Models",
+				"BaseURL":          utils.BaseURL(c),
 				"Version":          internal.PrintableVersion(),
 				"Models":           template.HTML(elements.ListModels(models, processingModels, galleryService)),
 				"Repositories":     appConfig.Galleries,
@@ -308,6 +311,7 @@ func RegisterUIRoutes(app *fiber.App,
 
 		summary := fiber.Map{
 			"Title":        "LocalAI - Chat with " + c.Params("model"),
+			"BaseURL":      utils.BaseURL(c),
 			"ModelsConfig": backendConfigs,
 			"Model":        c.Params("model"),
 			"Version":      internal.PrintableVersion(),
@@ -323,11 +327,12 @@ func RegisterUIRoutes(app *fiber.App,
 
 		if len(backendConfigs) == 0 {
 			// If no model is available redirect to the index which suggests how to install models
-			return c.Redirect("/")
+			return c.Redirect(utils.BaseURL(c))
 		}
 
 		summary := fiber.Map{
 			"Title":        "LocalAI - Talk",
+			"BaseURL":      utils.BaseURL(c),
 			"ModelsConfig": backendConfigs,
 			"Model":        backendConfigs[0],
 			"IsP2PEnabled": p2p.IsP2PEnabled(),
@@ -344,11 +349,12 @@ func RegisterUIRoutes(app *fiber.App,
 
 		if len(backendConfigs) == 0 {
 			// If no model is available redirect to the index which suggests how to install models
-			return c.Redirect("/")
+			return c.Redirect(utils.BaseURL(c))
 		}
 
 		summary := fiber.Map{
 			"Title":        "LocalAI - Chat with " + backendConfigs[0],
+			"BaseURL":      utils.BaseURL(c),
 			"ModelsConfig": backendConfigs,
 			"Model":        backendConfigs[0],
 			"Version":      internal.PrintableVersion(),
@@ -364,6 +370,7 @@ func RegisterUIRoutes(app *fiber.App,
 
 		summary := fiber.Map{
 			"Title":        "LocalAI - Generate images with " + c.Params("model"),
+			"BaseURL":      utils.BaseURL(c),
 			"ModelsConfig": backendConfigs,
 			"Model":        c.Params("model"),
 			"Version":      internal.PrintableVersion(),
@@ -380,11 +387,12 @@ func RegisterUIRoutes(app *fiber.App,
 
 		if len(backendConfigs) == 0 {
 			// If no model is available redirect to the index which suggests how to install models
-			return c.Redirect("/")
+			return c.Redirect(utils.BaseURL(c))
 		}
 
 		summary := fiber.Map{
 			"Title":        "LocalAI - Generate images with " + backendConfigs[0].Name,
+			"BaseURL":      utils.BaseURL(c),
 			"ModelsConfig": backendConfigs,
 			"Model":        backendConfigs[0].Name,
 			"Version":      internal.PrintableVersion(),
@@ -400,6 +408,7 @@ func RegisterUIRoutes(app *fiber.App,
 
 		summary := fiber.Map{
 			"Title":        "LocalAI - Generate images with " + c.Params("model"),
+			"BaseURL":      utils.BaseURL(c),
 			"ModelsConfig": backendConfigs,
 			"Model":        c.Params("model"),
 			"Version":      internal.PrintableVersion(),
@@ -416,11 +425,12 @@ func RegisterUIRoutes(app *fiber.App,
 
 		if len(backendConfigs) == 0 {
 			// If no model is available redirect to the index which suggests how to install models
-			return c.Redirect("/")
+			return c.Redirect(utils.BaseURL(c))
 		}
 
 		summary := fiber.Map{
 			"Title":        "LocalAI - Generate audio with " + backendConfigs[0].Name,
+			"BaseURL":      utils.BaseURL(c),
 			"ModelsConfig": backendConfigs,
 			"Model":        backendConfigs[0].Name,
 			"IsP2PEnabled": p2p.IsP2PEnabled(),

core/http/static/assets/font1.css

@@ -7,33 +7,33 @@ https://fonts.googleapis.com/css2?family=Inter:wght@400;600;700&family=Roboto:wg
   font-style: normal;
   font-weight: 400;
   font-display: swap;
-  src: url(/static/assets/UcCO3FwrK3iLTeHuS_fvQtMwCp50KnMw2boKoduKmMEVuLyfMZg.ttf) format('truetype');
+  src: url(./UcCO3FwrK3iLTeHuS_fvQtMwCp50KnMw2boKoduKmMEVuLyfMZg.ttf) format('truetype');
 }
 @font-face {
   font-family: 'Inter';
   font-style: normal;
   font-weight: 600;
   font-display: swap;
-  src: url(/static/assets/UcCO3FwrK3iLTeHuS_fvQtMwCp50KnMw2boKoduKmMEVuGKYMZg.ttf) format('truetype');
+  src: url(./UcCO3FwrK3iLTeHuS_fvQtMwCp50KnMw2boKoduKmMEVuGKYMZg.ttf) format('truetype');
 }
 @font-face {
   font-family: 'Inter';
   font-style: normal;
   font-weight: 700;
   font-display: swap;
-  src: url(/static/assets/UcCO3FwrK3iLTeHuS_fvQtMwCp50KnMw2boKoduKmMEVuFuYMZg.ttf) format('truetype');
+  src: url(./UcCO3FwrK3iLTeHuS_fvQtMwCp50KnMw2boKoduKmMEVuFuYMZg.ttf) format('truetype');
 }
 @font-face {
   font-family: 'Roboto';
   font-style: normal;
   font-weight: 400;
   font-display: swap;
-  src: url(/static/assets/KFOmCnqEu92Fr1Me5Q.ttf) format('truetype');
+  src: url(./KFOmCnqEu92Fr1Me5Q.ttf) format('truetype');
 }
 @font-face {
   font-family: 'Roboto';
   font-style: normal;
   font-weight: 500;
   font-display: swap;
-  src: url(/static/assets/KFOlCnqEu92Fr1MmEU9vAw.ttf) format('truetype');
+  src: url(./KFOlCnqEu92Fr1MmEU9vAw.ttf) format('truetype');
 }

core/http/static/assets/font2.css

@@ -7,33 +7,33 @@ https://fonts.googleapis.com/css?family=Roboto:300,400,500,700,900&display=swap
   font-style: normal;
   font-weight: 300;
   font-display: swap;
-  src: url(/static/assets//KFOlCnqEu92Fr1MmSU5fBBc9.ttf) format('truetype');
+  src: url(./KFOlCnqEu92Fr1MmSU5fBBc9.ttf) format('truetype');
 }
 @font-face {
   font-family: 'Roboto';
   font-style: normal;
   font-weight: 400;
   font-display: swap;
-  src: url(/static/assets//KFOmCnqEu92Fr1Mu4mxP.ttf) format('truetype');
+  src: url(./KFOmCnqEu92Fr1Mu4mxP.ttf) format('truetype');
 }
 @font-face {
   font-family: 'Roboto';
   font-style: normal;
   font-weight: 500;
   font-display: swap;
-  src: url(/static/assets//KFOlCnqEu92Fr1MmEU9fBBc9.ttf) format('truetype');
+  src: url(./KFOlCnqEu92Fr1MmEU9fBBc9.ttf) format('truetype');
 }
 @font-face {
   font-family: 'Roboto';
   font-style: normal;
   font-weight: 700;
   font-display: swap;
-  src: url(/static/assets//KFOlCnqEu92Fr1MmWUlfBBc9.ttf) format('truetype');
+  src: url(./KFOlCnqEu92Fr1MmWUlfBBc9.ttf) format('truetype');
 }
 @font-face {
   font-family: 'Roboto';
   font-style: normal;
   font-weight: 900;
   font-display: swap;
-  src: url(/static/assets//KFOlCnqEu92Fr1MmYUtfBBc9.ttf) format('truetype');
+  src: url(./KFOlCnqEu92Fr1MmYUtfBBc9.ttf) format('truetype');
 }

core/http/static/chat.js

@@ -143,7 +143,7 @@ function readInputImage() {
     // }
 
     // Source: https://stackoverflow.com/a/75751803/11386095
-    const response = await fetch("/v1/chat/completions", {
+    const response = await fetch("v1/chat/completions", {
       method: "POST",
       headers: {
         Authorization: `Bearer ${key}`,

core/http/static/image.js

@@ -48,7 +48,7 @@ async function promptDallE(key, input) {
   document.getElementById("input").disabled = true;
 
   const model = document.getElementById("image-model").value;
-  const response = await fetch("/v1/images/generations", {
+  const response = await fetch("v1/images/generations", {
     method: "POST",
     headers: {
       Authorization: `Bearer ${key}`,

core/http/static/talk.js

@@ -122,7 +122,7 @@ async function sendAudioToWhisper(audioBlob) {
     formData.append('model', getWhisperModel());
     API_KEY = localStorage.getItem("key");
 
-    const response = await fetch('/v1/audio/transcriptions', {
+    const response = await fetch('v1/audio/transcriptions', {
         method: 'POST',
         headers: {
             'Authorization': `Bearer ${API_KEY}`
@@ -139,7 +139,7 @@ async function sendTextToChatGPT(text) {
     conversationHistory.push({ role: "user", content: text });
     API_KEY = localStorage.getItem("key");
 
-    const response = await fetch('/v1/chat/completions', {
+    const response = await fetch('v1/chat/completions', {
         method: 'POST',
         headers: {
             'Authorization': `Bearer ${API_KEY}`,
@@ -163,7 +163,7 @@ async function sendTextToChatGPT(text) {
 async function getTextToSpeechAudio(text) {
     API_KEY = localStorage.getItem("key");
 
-    const response = await fetch('/v1/audio/speech', {
+    const response = await fetch('v1/audio/speech', {
         
         method: 'POST',
         headers: {

core/http/static/tts.js

@@ -19,7 +19,7 @@ async function tts(key, input) {
   document.getElementById("input").disabled = true;
 
   const model = document.getElementById("tts-model").value;
-  const response = await fetch("/tts", {
+  const response = await fetch("tts", {
     method: "POST",
     headers: {
       Authorization: `Bearer ${key}`,

core/http/utils/baseurl.go

@@ -0,0 +1,24 @@
+package utils
+
+import (
+	"strings"
+
+	"github.com/gofiber/fiber/v2"
+)
+
+// BaseURL returns the base URL for the given HTTP request context.
+// It takes into account that the app may be exposed by a reverse-proxy under a different protocol, host and path.
+// The returned URL is guaranteed to end with `/`.
+// The method should be used in conjunction with the StripPathPrefix middleware.
+func BaseURL(c *fiber.Ctx) string {
+	path := c.Path()
+	origPath := c.OriginalURL()
+
+	if path != origPath && strings.HasSuffix(origPath, path) {
+		pathPrefix := origPath[:len(origPath)-len(path)+1]
+
+		return c.BaseURL() + pathPrefix
+	}
+
+	return c.BaseURL() + "/"
+}

core/http/utils/baseurl_test.go

@@ -0,0 +1,48 @@
+package utils
+
+import (
+	"net/http/httptest"
+	"testing"
+
+	"github.com/gofiber/fiber/v2"
+	"github.com/stretchr/testify/require"
+)
+
+func TestBaseURL(t *testing.T) {
+	for _, tc := range []struct {
+		name      string
+		prefix    string
+		expectURL string
+	}{
+		{
+			name:      "without prefix",
+			prefix:    "/",
+			expectURL: "http://example.com/",
+		},
+		{
+			name:      "with prefix",
+			prefix:    "/myprefix/",
+			expectURL: "http://example.com/myprefix/",
+		},
+	} {
+		t.Run(tc.name, func(t *testing.T) {
+			app := fiber.New()
+			actualURL := ""
+
+			app.Get(tc.prefix+"hello/world", func(c *fiber.Ctx) error {
+				if tc.prefix != "/" {
+					c.Path("/hello/world")
+				}
+				actualURL = BaseURL(c)
+				return nil
+			})
+
+			req := httptest.NewRequest("GET", tc.prefix+"hello/world", nil)
+			resp, err := app.Test(req, -1)
+
+			require.NoError(t, err)
+			require.Equal(t, 200, resp.StatusCode, "response status code")
+			require.Equal(t, tc.expectURL, actualURL, "base URL")
+		})
+	}
+}

core/http/views/404.html

@@ -12,7 +12,7 @@
         <div class="header text-center py-12">
             <h1 class="text-5xl font-bold">Welcome to your LocalAI instance!</h1>
             <div class="mt-6">
-         <!--       <a href="/" aria-label="HomePage" alt="HomePage">           
+         <!--       <a href="./" aria-label="HomePage" alt="HomePage">
                     <img class="mx-auto w-1/4 h-auto" src="https://github.com/go-skynet/LocalAI/assets/2420543/0966aa2a-166e-4f99-a3e5-6c915fc997dd" alt="LocalAI Logo">            
                 </a>
             -->

core/http/views/chat.html

@@ -28,7 +28,7 @@ SOFTWARE.
 <!doctype html>
 <html lang="en">
   {{template "views/partials/head" .}}
-  <script defer src="/static/chat.js"></script>
+  <script defer src="static/chat.js"></script>
   <style>
     body {
         overflow: hidden; 
@@ -101,9 +101,9 @@ SOFTWARE.
         {{ $model:=.Model}}
         {{ range .ModelsConfig }}
         {{ if eq . $model }}
-        <option value="/chat/{{.}}" selected  class="bg-gray-700 text-white">{{.}}</option>
+        <option value="chat/{{.}}" selected  class="bg-gray-700 text-white">{{.}}</option>
         {{ else }}
-        <option value="/chat/{{.}}" class="bg-gray-700 text-white">{{.}}</option>
+        <option value="chat/{{.}}" class="bg-gray-700 text-white">{{.}}</option>
         {{ end }}
         {{ end }}
       </select>
@@ -142,7 +142,7 @@ SOFTWARE.
       <div id="loader" class="my-2 loader" style="display: none;"></div>
       <input id="chat-model" type="hidden" value="{{.Model}}">
       <input id="input_image" type="file" style="display: none;" @change="fileName = $event.target.files[0].name">
-      <form id="prompt" action="/chat/{{.Model}}" method="get" @submit.prevent="submitPrompt">
+      <form id="prompt" action="chat/{{.Model}}" method="get" @submit.prevent="submitPrompt">
           <div class="relative w-full">
               <textarea
                   id="input"

core/http/views/explorer.html

@@ -370,7 +370,7 @@
                 }
             }
         </script>
-        <script src="/static/p2panimation.js"></script>
+        <script src="static/p2panimation.js"></script>
 
         {{template "views/partials/footer" .}}
     </div>

core/http/views/index.html

@@ -20,7 +20,7 @@
             {{template "views/partials/inprogress" .}}
             {{ if eq (len .ModelsConfig) 0 }}
             <h2 class="text-center text-3xl font-semibold text-gray-100"> <i class="text-yellow-200 ml-2 fa-solid fa-triangle-exclamation animate-pulse"></i> Ouch! seems you don't have any models installed from the LocalAI gallery!</h2>
-            <p class="text-center mt-4 text-xl">..install something from the <a class="text-gray-400 hover:text-white ml-1 px-3 py-2 rounded" href="/browse">🖼️ Gallery</a> or check the <a href="https://localai.io/basics/getting_started/" class="text-gray-400 hover:text-white ml-1 px-3 py-2 rounded"> <i class="fa-solid fa-book"></i> Getting started documentation </a></p>
+            <p class="text-center mt-4 text-xl">..install something from the <a class="text-gray-400 hover:text-white ml-1 px-3 py-2 rounded" href="browse">🖼️ Gallery</a> or check the <a href="https://localai.io/basics/getting_started/" class="text-gray-400 hover:text-white ml-1 px-3 py-2 rounded"> <i class="fa-solid fa-book"></i> Getting started documentation </a></p>
 
             {{ if ne (len .Models) 0 }}
             <hr class="my-4">
@@ -66,7 +66,7 @@
                         {{ end }}
                     </td>
                     <td class="px-4 py-3 font-bold">
-                        <p class="font-bold text-white flex items-center"><i class="fas fa-brain pr-2"></i><a href="/browse?term={{.Name}}">{{.Name}}</a></p>
+                        <p class="font-bold text-white flex items-center"><i class="fas fa-brain pr-2"></i><a href="browse?term={{.Name}}">{{.Name}}</a></p>
                     </td>
                     <td class="px-4 py-3 font-bold">
                         {{ if .Backend }}
@@ -84,7 +84,7 @@
                     <td class="px-4 py-3">
                         <button
                             class="float-right inline-block rounded bg-red-800 px-6 pb-2.5 mb-3 pt-2.5 text-xs font-medium uppercase leading-normal text-white shadow-primary-3 transition duration-150 ease-in-out hover:bg-red-accent-300 hover:shadow-red-2 focus:bg-red-accent-300 focus:shadow-primary-2 focus:outline-none focus:ring-0 active:bg-red-600 active:shadow-primary-2 dark:shadow-black/30 dark:hover:shadow-dark-strong dark:focus:shadow-dark-strong dark:active:shadow-dark-strong"
-                            data-twe-ripple-color="light" data-twe-ripple-init="" hx-confirm="Are you sure you wish to delete the model?" hx-post="/browse/delete/model/{{.Name}}" hx-swap="outerHTML"><i class="fa-solid fa-cancel pr-2"></i>Delete</button>
+                            data-twe-ripple-color="light" data-twe-ripple-init="" hx-confirm="Are you sure you wish to delete the model?" hx-post="browse/delete/model/{{.Name}}" hx-swap="outerHTML"><i class="fa-solid fa-cancel pr-2"></i>Delete</button>
                     </td>
                 {{ end }}
                 {{ range .Models }}

core/http/views/login.html

@@ -4,6 +4,8 @@
     <meta charset="UTF-8">
     <meta name="viewport" content="width=device-width, initial-scale=1.0">
     <title>Open Authenticated Website</title>
+    <base href="{{.BaseURL}}" />
+    <link rel="icon" type="image/x-icon" href="favicon.ico" />
 </head>
 <body>
     <h1>Authorization is required</h1>

core/http/views/models.html

@@ -16,38 +16,38 @@
 
             <div class="text-center font-semibold text-gray-100">
                 <h2>Filter by type:</h2>
-                <button  hx-post="/browse/search/models"
+                <button  hx-post="browse/search/models"
                     class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
                     hx-target="#search-results" 
                     hx-vals='{"search": "tts"}'
                 hx-indicator=".htmx-indicator" >TTS</button> 
-                <button  hx-post="/browse/search/models" 
+                <button  hx-post="browse/search/models" 
                     class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
                     hx-target="#search-results" 
                     hx-vals='{"search": "stablediffusion"}'
                 hx-indicator=".htmx-indicator" >Image generation</button> 
-                <button  hx-post="/browse/search/models" \
+                <button  hx-post="browse/search/models" \
                     class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
                     hx-target="#search-results" 
                     hx-vals='{"search": "llm"}'
                 hx-indicator=".htmx-indicator" >Text generation</button> 
-                <button  hx-post="/browse/search/models" 
+                <button  hx-post="browse/search/models" 
                     class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
                     hx-target="#search-results" 
                     hx-vals='{"search": "multimodal"}'
                 hx-indicator=".htmx-indicator" >Multimodal</button> 
-                <button  hx-post="/browse/search/models" 
+                <button  hx-post="browse/search/models" 
                     class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
                     hx-target="#search-results" 
                     hx-vals='{"search": "embedding"}'
                 hx-indicator=".htmx-indicator" >Embeddings</button>
-                <button  hx-post="/browse/search/models"
+                <button  hx-post="browse/search/models"
                     class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
                     hx-target="#search-results" 
                     hx-vals='{"search": "rerank"}'
                 hx-indicator=".htmx-indicator" >Rerankers</button> 
                 <button  
-                    hx-post="/browse/search/models"
+                    hx-post="browse/search/models"
                     class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
                     hx-target="#search-results" 
                     hx-vals='{"search": "whisper"}'
@@ -57,7 +57,7 @@
             <div class="text-center text-xs font-semibold text-gray-100">
                 Filter by tags:
                 {{ range .AllTags }}
-                    <button  hx-post="/browse/search/models" class="text-blue-500" hx-target="#search-results" 
+                    <button  hx-post="browse/search/models" class="text-blue-500" hx-target="#search-results" 
                     hx-vals='{"search": "{{.}}"}'
                     hx-indicator=".htmx-indicator" >{{.}}</button> 
                 {{ end }}
@@ -69,7 +69,7 @@
 
             <input class="form-control appearance-none block w-full mt-5 px-3 py-2 text-base font-normal text-gray-300 pb-2 mb-5 bg-gray-800 bg-clip-padding border border-solid border-gray-600 rounded transition ease-in-out m-0 focus:text-gray-300 focus:bg-gray-900 focus:border-blue-500 focus:outline-none" type="search" 
                 name="search" placeholder="Begin Typing To Search models..." 
-                hx-post="/browse/search/models" 
+                hx-post="browse/search/models" 
                 hx-trigger="input changed delay:500ms, search" 
                 hx-target="#search-results" 
                 hx-indicator=".htmx-indicator">

core/http/views/p2p.html

@@ -48,11 +48,11 @@
             <!-- Federation Box -->
             <div class="bg-gray-800 p-6 rounded-lg shadow-lg mb-12 text-left">
 
-                <p class="text-xl font-semibold text-gray-200"> <i class="text-gray-200 fa-solid fa-circle-nodes"></i> Federated Nodes: <span hx-get="/p2p/ui/workers-federation-stats" hx-trigger="every 1s"></span> </p>
+                <p class="text-xl font-semibold text-gray-200"> <i class="text-gray-200 fa-solid fa-circle-nodes"></i> Federated Nodes: <span hx-get="p2p/ui/workers-federation-stats" hx-trigger="every 1s"></span> </p>
                 <p class="mb-4">You can start LocalAI in federated mode to share your instance, or start the federated server to balance requests between nodes of the federation.</p>
 
                 <div class="grid grid-cols-1 sm:grid-cols-2 md:grid-cols-3 gap-4 mb-12">
-                    <div hx-get="/p2p/ui/workers-federation" hx-trigger="every 1s"></div>
+                    <div hx-get="p2p/ui/workers-federation" hx-trigger="every 1s"></div>
                 </div>
 
                 <hr class="border-gray-700 mb-12">
@@ -123,11 +123,11 @@
 
             <div class="bg-gray-800 p-6 rounded-lg shadow-lg mb-12 text-left">
 
-                <p class="text-xl font-semibold text-gray-200"> <i class="text-gray-200 fa-solid fa-circle-nodes"></i> Workers (llama.cpp): <span hx-get="/p2p/ui/workers-stats" hx-trigger="every 1s"></span> </p>
+                <p class="text-xl font-semibold text-gray-200"> <i class="text-gray-200 fa-solid fa-circle-nodes"></i> Workers (llama.cpp): <span hx-get="p2p/ui/workers-stats" hx-trigger="every 1s"></span> </p>
                 <p class="mb-4">You can start llama.cpp workers to distribute weights between the workers and offload part of the computation. To start a new worker, you can use the CLI or Docker.</p>
 
                 <div class="grid grid-cols-1 sm:grid-cols-2 md:grid-cols-3 gap-4 mb-12">
-                    <div hx-get="/p2p/ui/workers" hx-trigger="every 1s"></div>
+                    <div hx-get="p2p/ui/workers" hx-trigger="every 1s"></div>
                 </div>
                 <hr class="border-gray-700 mb-12">
 
@@ -177,7 +177,7 @@
 
     {{template "views/partials/footer" .}}
 </div>
-<script src="/static/p2panimation.js"></script>
+<script src="static/p2panimation.js"></script>
 <style>
     .token {
         word-break: break-all;

core/http/views/partials/footer.html

@@ -2,4 +2,4 @@
     LocalAI Version {{.Version}}<br>
     <a href='https://github.com/mudler/LocalAI' class="text-blue-400 hover:text-blue-600" target="_blank">LocalAI</a> © 2023-2024 <a href='https://mudler.pm' class="text-blue-400 hover:text-blue-600" target="_blank">Ettore Di Giacinto</a>
 </footer>
-<script src="/static/assets/tw-elements.js"></script>
+<script src="static/assets/tw-elements.js"></script>

core/http/views/partials/head.html

@@ -2,33 +2,35 @@
     <meta charset="UTF-8">
     <meta name="viewport" content="width=device-width, initial-scale=1.0">
     <title>{{.Title}}</title>
+    <base href="{{.BaseURL}}" />
+    <link rel="icon" type="image/x-icon" href="favicon.ico" />
     <link
     rel="stylesheet"
-    href="/static/assets/highlightjs.css"
+    href="static/assets/highlightjs.css"
   />
-  <script defer src="/static/assets/highlightjs.js"></script>
+  <script defer src="static/assets/highlightjs.js"></script>
     <script
     defer
-    src="/static/assets/alpine.js"
+    src="static/assets/alpine.js"
   ></script>
   <script
     defer
-    src="/static/assets/marked.js"
+    src="static/assets/marked.js"
   ></script>
   <script
     defer
-    src="/static/assets/purify.js"
+    src="static/assets/purify.js"
   ></script>
 
-  <link href="/static/general.css" rel="stylesheet" />
+  <link href="static/general.css" rel="stylesheet" />
-    <link href="/static/assets/font1.css" rel="stylesheet">
+    <link href="static/assets/font1.css" rel="stylesheet">
     <link
-    href="/static/assets/font2.css"
+    href="static/assets/font2.css"
     rel="stylesheet" />
   <link
     rel="stylesheet"
-    href="/static/assets/tw-elements.css" />
+    href="static/assets/tw-elements.css" />
-  <script src="/static/assets/tailwindcss.js"></script>
+  <script src="static/assets/tailwindcss.js"></script>
   <script>
     tailwind.config = {
       darkMode: "class",
@@ -54,11 +56,11 @@
       });
     }
   </script>
-  <link href="/static/assets/fontawesome/css/fontawesome.css" rel="stylesheet" />
+  <link href="static/assets/fontawesome/css/fontawesome.css" rel="stylesheet" />
-  <link href="/static/assets/fontawesome/css/brands.css" rel="stylesheet" />
+  <link href="static/assets/fontawesome/css/brands.css" rel="stylesheet" />
-  <link href="/static/assets/fontawesome/css/solid.css" rel="stylesheet" />
+  <link href="static/assets/fontawesome/css/solid.css" rel="stylesheet" />
-  <script src="/static/assets/flowbite.min.js"></script>
+  <script src="static/assets/flowbite.min.js"></script>
-  <script src="/static/assets/htmx.js" crossorigin="anonymous"></script>
+  <script src="static/assets/htmx.js" crossorigin="anonymous"></script>
   <!-- P2P Animation START -->
   <style>
     .animation-container {

core/http/views/partials/inprogress.html

@@ -17,13 +17,13 @@
 
       <div class="flex items-center justify-between bg-slate-600 p-2 mb-2 rounded-md">
          <div class="flex items center">
-             <span class="text-gray-300"><a href="/browse?term={{$parts._1}}"
+             <span class="text-gray-300"><a href="browse?term={{$parts._1}}"
                  class="text-white-500 inline-block bg-blue-200 rounded-full px-3 py-1 text-sm font-semibold text-gray-700 mr-2 mb-2 hover:bg-gray-300 hover:shadow-gray-2"
                  >{{$modelName}}</a> {{if $repository}} (from the '{{$repository}}' repository) {{end}}</span>
          </div>
-         <div hx-get="/browse/job/{{$value}}" hx-swap="outerHTML" hx-target="this" hx-trigger="done">
+         <div hx-get="browse/job/{{$value}}" hx-swap="outerHTML" hx-target="this" hx-trigger="done">
              <h3 role="status" id="pblabel" >{{$op}}
-                 <div hx-get="/browse/job/progress/{{$value}}" hx-trigger="every 600ms" 
+                 <div hx-get="browse/job/progress/{{$value}}" hx-trigger="every 600ms" 
                  hx-target="this"
                  hx-swap="innerHTML"  ></div></h3>
          </div>     

core/http/views/partials/navbar.html

@@ -3,8 +3,8 @@
         <div class="flex items-center justify-between">
             <div class="flex items-center">
                 <!-- Logo Image: Replace 'logo_url_here' with your actual logo URL -->
-                <a href="/" class="text-white text-xl font-bold"><img src="https://github.com/go-skynet/LocalAI/assets/2420543/0966aa2a-166e-4f99-a3e5-6c915fc997dd" alt="LocalAI Logo" class="h-10 mr-3 border-2 border-gray-300 shadow rounded"></a>
+                <a href="./" class="text-white text-xl font-bold"><img src="https://github.com/go-skynet/LocalAI/assets/2420543/0966aa2a-166e-4f99-a3e5-6c915fc997dd" alt="LocalAI Logo" class="h-10 mr-3 border-2 border-gray-300 shadow rounded"></a>
-                <a href="/" class="text-white text-xl font-bold">LocalAI</a>
+                <a href="./" class="text-white text-xl font-bold">LocalAI</a>
             </div>
             <!-- Menu button for small screens -->
             <div class="lg:hidden">
@@ -14,33 +14,33 @@
             </div>
             <!-- Navigation links -->
             <div class="hidden lg:flex lg:items-center lg:justify-end lg:flex-1 lg:w-0">
-                <a href="/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-home pr-2"></i>Home</a>
+                <a href="./" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-home pr-2"></i>Home</a>
                 <a href="https://localai.io" class="text-gray-400 hover:text-white px-3 py-2 rounded" target="_blank" ><i class="fas fa-book-reader pr-2"></i> Documentation</a>
-                <a href="/browse/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-brain pr-2"></i> Models</a>
+                <a href="browse/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-brain pr-2"></i> Models</a>
-                <a href="/chat/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fa-solid fa-comments pr-2"></i> Chat</a>
+                <a href="chat/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fa-solid fa-comments pr-2"></i> Chat</a>
-                <a href="/text2image/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-image pr-2"></i> Generate images</a>
+                <a href="text2image/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-image pr-2"></i> Generate images</a>
-                <a href="/tts/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fa-solid fa-music pr-2"></i> TTS </a>
+                <a href="tts/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fa-solid fa-music pr-2"></i> TTS </a>
-                <a href="/talk/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fa-solid fa-phone pr-2"></i> Talk </a>
+                <a href="talk/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fa-solid fa-phone pr-2"></i> Talk </a>
                 {{ if .IsP2PEnabled }}
-                <a href="/p2p/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fa-solid fa-circle-nodes"></i> Swarm </a>
+                <a href="p2p/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fa-solid fa-circle-nodes"></i> Swarm </a>
                 {{ end }}
-                <a href="/swagger/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-code pr-2"></i> API</a>
+                <a href="swagger/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-code pr-2"></i> API</a>
             </div>
         </div>
         <!-- Collapsible menu for small screens -->
         <div class="hidden lg:hidden" id="mobile-menu">
             <div class="pt-4 pb-3 border-t border-gray-700">
-                <a href="/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fas fa-home pr-2"></i>Home</a>
+                <a href="./" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fas fa-home pr-2"></i>Home</a>
                 <a href="https://localai.io" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1" target="_blank" ><i class="fas fa-book-reader pr-2"></i> Documentation</a>
-                <a href="/browse/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fas fa-brain pr-2"></i> Models</a>
+                <a href="browse/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fas fa-brain pr-2"></i> Models</a>
-                <a href="/chat/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fa-solid fa-comments pr-2"></i> Chat</a>
+                <a href="chat/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fa-solid fa-comments pr-2"></i> Chat</a>
-                <a href="/text2image/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fas fa-image pr-2"></i> Generate images</a>
+                <a href="text2image/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fas fa-image pr-2"></i> Generate images</a>
-                <a href="/tts/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fa-solid fa-music pr-2"></i> TTS </a>
+                <a href="tts/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fa-solid fa-music pr-2"></i> TTS </a>
-                <a href="/talk/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fa-solid fa-phone pr-2"></i> Talk </a>
+                <a href="talk/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fa-solid fa-phone pr-2"></i> Talk </a>
                 {{ if .IsP2PEnabled }}
-                <a href="/p2p/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fa-solid fa-circle-nodes"></i> Swarm </a>
+                <a href="p2p/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fa-solid fa-circle-nodes"></i> Swarm </a>
                 {{ end }}
-                <a href="/swagger/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fas fa-code pr-2"></i> API</a>
+                <a href="swagger/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fas fa-code pr-2"></i> API</a>
             </div>
         </div>
     </div>

core/http/views/partials/navbar_explorer.html

@@ -3,8 +3,8 @@
         <div class="flex items-center justify-between">
             <div class="flex items-center">
                 <!-- Logo Image: Replace 'logo_url_here' with your actual logo URL -->
-                <a href="/" class="text-white text-xl font-bold"><img src="https://github.com/go-skynet/LocalAI/assets/2420543/0966aa2a-166e-4f99-a3e5-6c915fc997dd" alt="LocalAI Logo" class="h-10 mr-3 border-2 border-gray-300 shadow rounded"></a>
+                <a href="./" class="text-white text-xl font-bold"><img src="https://github.com/go-skynet/LocalAI/assets/2420543/0966aa2a-166e-4f99-a3e5-6c915fc997dd" alt="LocalAI Logo" class="h-10 mr-3 border-2 border-gray-300 shadow rounded"></a>
-                <a href="/" class="text-white text-xl font-bold">LocalAI</a>
+                <a href="./" class="text-white text-xl font-bold">LocalAI</a>
             </div>
             <!-- Menu button for small screens -->
             <div class="lg:hidden">
@@ -14,7 +14,7 @@
             </div>
             <!-- Navigation links -->
             <div class="hidden lg:flex lg:items-center lg:justify-end lg:flex-1 lg:w-0">
-                <a href="/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-home pr-2"></i>Home</a>
+                <a href="./" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-home pr-2"></i>Home</a>
                 <a href="https://localai.io" class="text-gray-400 hover:text-white px-3 py-2 rounded" target="_blank" ><i class="fas fa-book-reader pr-2"></i> Documentation</a>
                 <a href="https://models.localai.io/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-brain pr-2"></i> Models</a>
             </div>
@@ -22,7 +22,7 @@
         <!-- Collapsible menu for small screens -->
         <div class="hidden lg:hidden" id="mobile-menu">
             <div class="pt-4 pb-3 border-t border-gray-700">
-                <a href="/" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fas fa-home pr-2"></i>Home</a>
+                <a href="./" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1"><i class="fas fa-home pr-2"></i>Home</a>
                 <a href="https://localai.io" class="block text-gray-400 hover:text-white px-3 py-2 rounded mt-1" target="_blank" ><i class="fas fa-book-reader pr-2"></i> Documentation</a>
                 <a href="https://models.localai.io/" class="text-gray-400 hover:text-white px-3 py-2 rounded"><i class="fas fa-brain pr-2"></i> Models</a>
             </div>

core/http/views/talk.html

@@ -1,7 +1,7 @@
 <!doctype html>
 <html lang="en">
   {{template "views/partials/head" .}}
-  <script defer src="/static/talk.js"></script>
+  <script defer src="static/talk.js"></script>
   <style>
     body {
         overflow: hidden; 

core/http/views/text2image.html

@@ -1,7 +1,7 @@
 <!DOCTYPE html>
 <html lang="en">
 {{template "views/partials/head" .}}
-<script defer src="/static/image.js"></script>
+<script defer src="static/image.js"></script>
 
 <body class="bg-gray-900 text-gray-200">
 <div class="flex flex-col min-h-screen">
@@ -50,9 +50,9 @@
                 {{ $model:=.Model}}
                 {{ range .ModelsConfig }}
                 {{ if eq .Name $model }}
-                <option value="/text2image/{{.Name}}" selected class="bg-gray-700 text-white">{{.Name}}</option>
+                <option value="text2image/{{.Name}}" selected class="bg-gray-700 text-white">{{.Name}}</option>
                 {{ else }}
-                <option value="/text2image/{{.Name}}" class="bg-gray-700 text-white">{{.Name}}</option>
+                <option value="text2image/{{.Name}}" class="bg-gray-700 text-white">{{.Name}}</option>
                 {{ end }}
                 {{ end }}
               </select>
@@ -62,7 +62,7 @@
 
             <div class="mt-12">
               <input id="image-model" type="hidden" value="{{.Model}}">
-              <form id="genimage" action="/text2image/{{.Model}}" method="get">
+              <form id="genimage" action="text2image/{{.Model}}" method="get">
                 <input
                   type="text"
                   id="input"

core/http/views/tts.html

@@ -1,7 +1,7 @@
 <!DOCTYPE html>
 <html lang="en">
 {{template "views/partials/head" .}}
-<script defer src="/static/tts.js"></script>
+<script defer src="static/tts.js"></script>
 
 <body class="bg-gray-900 text-gray-200">
 <div class="flex flex-col min-h-screen">
@@ -47,9 +47,9 @@
                 {{ $model:=.Model}}
                 {{ range .ModelsConfig }}
                 {{ if eq .Name $model }}
-                <option value="/tts/{{.Name}}" selected class="bg-gray-700 text-white">{{.Name}}</option>
+                <option value="tts/{{.Name}}" selected class="bg-gray-700 text-white">{{.Name}}</option>
                 {{ else }}
-                <option value="/tts/{{.Name}}" class="bg-gray-700 text-white">{{.Name}}</option>
+                <option value="tts/{{.Name}}" class="bg-gray-700 text-white">{{.Name}}</option>
                 {{ end }}
                 {{ end }}
               </select>
@@ -59,7 +59,7 @@
 
             <div class="mt-12">
               <input id="tts-model" type="hidden" value="{{.Model}}">
-              <form id="tts" action="/tts/{{.Model}}" method="get">
+              <form id="tts" action="tts/{{.Model}}" method="get">
                 <input
                   type="text"
                   id="input"

docs/content/docs/reference/compatibility-table.md

@@ -16,8 +16,8 @@ LocalAI will attempt to automatically load models which are not explicitly confi
 
 | Backend and Bindings                                                             | Compatible models     | Completion/Chat endpoint | Capability | Embeddings support                | Token stream support | Acceleration |
 |----------------------------------------------------------------------------------|-----------------------|--------------------------|---------------------------|-----------------------------------|----------------------|--------------|
-| [llama.cpp]({{%relref "docs/features/text-generation#llama.cpp" %}})        | LLama, Mamba, RWKV, Falcon, Starcoder, GPT-2, [and many others](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#description) | yes                      | GPT and Functions                        | yes** | yes                  | CUDA, openCL, cuBLAS, Metal |
+| [llama.cpp]({{%relref "docs/features/text-generation#llama.cpp" %}})        | LLama, Mamba, RWKV, Falcon, Starcoder, GPT-2, [and many others](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#description) | yes                      | GPT and Functions                        | yes | yes                  | CUDA, openCL, cuBLAS, Metal |
-| [llama.cpp's ggml model (backward compatibility with old format, before GGUF)](https://github.com/ggerganov/llama.cpp) ([binding](https://github.com/go-skynet/go-llama.cpp))  | LLama, GPT-2, [and many others](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#description) | yes                      | GPT and Functions                        | yes** | yes                  | CUDA, openCL, cuBLAS, Metal |
+| [llama.cpp's ggml model (backward compatibility with old format, before GGUF)](https://github.com/ggerganov/llama.cpp) ([binding](https://github.com/go-skynet/go-llama.cpp))  | LLama, GPT-2, [and many others](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#description) | yes                      | GPT and Functions                        | yes | yes                  | CUDA, openCL, cuBLAS, Metal |
 | [whisper](https://github.com/ggerganov/whisper.cpp)         | whisper               | no                       | Audio                 | no                                | no                   | N/A |
 | [stablediffusion](https://github.com/EdVince/Stable-Diffusion-NCNN) ([binding](https://github.com/mudler/go-stable-diffusion))        | stablediffusion               | no                       | Image                 | no                                | no                   | N/A |
 | [langchain-huggingface](https://github.com/tmc/langchaingo)                                                                    | Any text generators available on HuggingFace through API | yes                      | GPT                        | no                                | no                   | N/A |
@@ -37,14 +37,11 @@ LocalAI will attempt to automatically load models which are not explicitly confi
 | `openvoice` | Open voice    | no                       | Audio generation and Voice cloning    | no                               | no                   | CPU/CUDA |
 | `parler-tts` | Open voice    | no                       | Audio generation and Voice cloning    | no                               | no                   | CPU/CUDA |
 | [rerankers](https://github.com/AnswerDotAI/rerankers) | Reranking API    | no                       | Reranking   | no                               | no                   | CPU/CUDA |
-| `transformers` | Various GPTs and quantization formats | yes                      | GPT, embeddings            | yes | yes****                  | CPU/CUDA/XPU |
+| `transformers` | Various GPTs and quantization formats | yes                      | GPT, embeddings            | yes | yes*                  | CPU/CUDA/XPU |
 | [bark-cpp](https://github.com/PABannier/bark.cpp)        | bark               | no                       | Audio-Only                 | no                                | no                   | yes |
 | [stablediffusion-cpp](https://github.com/leejet/stable-diffusion.cpp)         | stablediffusion-1, stablediffusion-2, stablediffusion-3, flux, PhotoMaker               | no                       | Image                 | no                                | no                   | N/A |
 | [silero-vad](https://github.com/snakers4/silero-vad) with [Golang bindings](https://github.com/streamer45/silero-vad-go) | Silero VAD    | no                       | Voice Activity Detection    | no                               | no                   | CPU |
 
 Note: any backend name listed above can be used in the `backend` field of the model configuration file (See [the advanced section]({{%relref "docs/advanced" %}})).
 
-- \* 7b ONLY
+- \* Only for CUDA and OpenVINO CPU/XPU acceleration.
-- ** doesn't seem to be accurate
-- *** 7b and 40b with the `ggccv` format, for instance: https://huggingface.co/TheBloke/WizardLM-Uncensored-Falcon-40B-GGML
-- **** Only for CUDA and OpenVINO CPU/XPU acceleration.

docs/themes/hugo-theme-relearn

@@ -1 +1 @@
-Subproject commit d25f856477223170b0de0b284252aa54b3e6255b
+Subproject commit 80e448e5bdaa92c87ee0d0d86f1125c8606ebf5f

gallery/index.yaml

@@ -1069,6 +1069,56 @@
     - filename: Llama-Deepsync-3B.Q4_K_M.gguf
       sha256: f11c4d9b10a732845d8e64dc9badfcbb7d94053bc5fe11f89bb8e99ed557f711
       uri: huggingface://prithivMLmods/Llama-Deepsync-3B-GGUF/Llama-Deepsync-3B.Q4_K_M.gguf
+- !!merge <<: *llama32
+  name: "dolphin3.0-llama3.2-1b"
+  url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
+  icon: https://cdn-uploads.huggingface.co/production/uploads/63111b2d88942700629f5771/cNCs1TBD3FelWCJGkZ3cd.png
+  urls:
+    - https://huggingface.co/cognitivecomputations/Dolphin3.0-Llama3.2-1B
+    - https://huggingface.co/bartowski/Dolphin3.0-Llama3.2-1B-GGUF
+  description: |
+    Dolphin 3.0 is the next generation of the Dolphin series of instruct-tuned models. Designed to be the ultimate general purpose local model, enabling coding, math, agentic, function calling, and general use cases.
+
+    Dolphin aims to be a general purpose model, similar to the models behind ChatGPT, Claude, Gemini. But these models present problems for businesses seeking to include AI in their products.
+
+        They maintain control of the system prompt, deprecating and changing things as they wish, often causing software to break.
+        They maintain control of the model versions, sometimes changing things silently, or deprecating older models that your business relies on.
+        They maintain control of the alignment, and in particular the alignment is one-size-fits all, not tailored to the application.
+        They can see all your queries and they can potentially use that data in ways you wouldn't want. Dolphin, in contrast, is steerable and gives control to the system owner. You set the system prompt. You decide the alignment. You have control of your data. Dolphin does not impose its ethics or guidelines on you. You are the one who decides the guidelines.
+
+    Dolphin belongs to YOU, it is your tool, an extension of your will. Just as you are personally responsible for what you do with a knife, gun, fire, car, or the internet, you are the creator and originator of any content you generate with Dolphin.
+  overrides:
+    parameters:
+      model: Dolphin3.0-Llama3.2-1B-Q4_K_M.gguf
+  files:
+    - filename: Dolphin3.0-Llama3.2-1B-Q4_K_M.gguf
+      sha256: 7ed39ee0638e18d3e47bf12e60e917c792ca5332606a72bd1882ab1f62a13a7a
+      uri: huggingface://bartowski/Dolphin3.0-Llama3.2-1B-GGUF/Dolphin3.0-Llama3.2-1B-Q4_K_M.gguf
+- !!merge <<: *llama32
+  name: "dolphin3.0-llama3.2-3b"
+  icon: https://cdn-uploads.huggingface.co/production/uploads/63111b2d88942700629f5771/cNCs1TBD3FelWCJGkZ3cd.png
+  url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
+  urls:
+    - https://huggingface.co/cognitivecomputations/Dolphin3.0-Llama3.2-3B
+    - https://huggingface.co/bartowski/Dolphin3.0-Llama3.2-3B-GGUF
+  description: |
+    Dolphin 3.0 is the next generation of the Dolphin series of instruct-tuned models. Designed to be the ultimate general purpose local model, enabling coding, math, agentic, function calling, and general use cases.
+
+    Dolphin aims to be a general purpose model, similar to the models behind ChatGPT, Claude, Gemini. But these models present problems for businesses seeking to include AI in their products.
+
+        They maintain control of the system prompt, deprecating and changing things as they wish, often causing software to break.
+        They maintain control of the model versions, sometimes changing things silently, or deprecating older models that your business relies on.
+        They maintain control of the alignment, and in particular the alignment is one-size-fits all, not tailored to the application.
+        They can see all your queries and they can potentially use that data in ways you wouldn't want. Dolphin, in contrast, is steerable and gives control to the system owner. You set the system prompt. You decide the alignment. You have control of your data. Dolphin does not impose its ethics or guidelines on you. You are the one who decides the guidelines.
+
+    Dolphin belongs to YOU, it is your tool, an extension of your will. Just as you are personally responsible for what you do with a knife, gun, fire, car, or the internet, you are the creator and originator of any content you generate with Dolphin.
+  overrides:
+    parameters:
+      model: Dolphin3.0-Llama3.2-3B-Q4_K_M.gguf
+  files:
+    - filename: Dolphin3.0-Llama3.2-3B-Q4_K_M.gguf
+      sha256: 5d6d02eeefa1ab5dbf23f97afdf5c2c95ad3d946dc3b6e9ab72e6c1637d54177
+      uri: huggingface://bartowski/Dolphin3.0-Llama3.2-3B-GGUF/Dolphin3.0-Llama3.2-3B-Q4_K_M.gguf
 - &qwen25
   ## Qwen2.5
   name: "qwen2.5-14b-instruct"
@@ -2662,6 +2712,119 @@
     - filename: lwd-Mirau-RP-Q4_K_M-imat.gguf
       sha256: 22ff461e9034b9ebded07b2a9d3d88c2f75359d5c069ebb3ee4e9c6ec5c45cf8
       uri: huggingface://Lewdiculous/experimental-lwd-Mirau-RP-14B-GGUF-IQ-Imatrix/lwd-Mirau-RP-Q4_K_M-imat.gguf
+- !!merge <<: *qwen25
+  name: "32b-qwen2.5-kunou-v1"
+  icon: https://huggingface.co/Sao10K/72B-Qwen2.5-Kunou-v1/resolve/main/knn.png
+  urls:
+    - https://huggingface.co/Sao10K/32B-Qwen2.5-Kunou-v1
+    - https://huggingface.co/bartowski/32B-Qwen2.5-Kunou-v1-GGUF
+  description: |
+    I do not really have anything planned for this model other than it being a generalist, and Roleplay Model? It was just something made and planned in minutes.
+    Same with the 14B and 72B version.
+    Kunou's the name of an OC I worked on for a couple of years, for a... fanfic. mmm...
+    A kind-of successor to L3-70B-Euryale-v2.2 in all but name? I'm keeping Stheno/Euryale lineage to Llama series for now.
+    I had a version made on top of Nemotron, a supposed Euryale 2.4 but that flopped hard, it was not my cup of tea.
+    This version is basically a better, more cleaned up Dataset used on Euryale and Stheno.
+  overrides:
+    parameters:
+      model: 32B-Qwen2.5-Kunou-v1-Q4_K_M.gguf
+  files:
+    - filename: 32B-Qwen2.5-Kunou-v1-Q4_K_M.gguf
+      sha256: b8910172b74d03c3463ac301589f54b96e54f61c67531fb6b523ecfe923aaffb
+      uri: huggingface://bartowski/32B-Qwen2.5-Kunou-v1-GGUF/32B-Qwen2.5-Kunou-v1-Q4_K_M.gguf
+- !!merge <<: *qwen25
+  name: "14b-qwen2.5-kunou-v1"
+  urls:
+    - https://huggingface.co/Sao10K/14B-Qwen2.5-Kunou-v1
+    - https://huggingface.co/DevQuasar/Sao10K.14B-Qwen2.5-Kunou-v1-GGUF
+  description: |
+    I do not really have anything planned for this model other than it being a generalist, and Roleplay Model? It was just something made and planned in minutes.
+    This is the little sister variant, the small 14B version.
+    Kunou's the name of an OC I worked on for a couple of years, for a... fanfic. mmm...
+
+    A kind-of successor to my smaller model series. It works pretty nicely I think?
+    This version is basically a better, more cleaned up Dataset used on Euryale and Stheno.
+  overrides:
+    parameters:
+      model: Sao10K.14B-Qwen2.5-Kunou-v1.Q4_K_M.gguf
+  files:
+    - filename: Sao10K.14B-Qwen2.5-Kunou-v1.Q4_K_M.gguf
+      sha256: 7b7af50076e15c305a2a1bed7ad766dc6deb61eef3c2e6a40d4c94ad45623845
+      uri: huggingface://DevQuasar/Sao10K.14B-Qwen2.5-Kunou-v1-GGUF/Sao10K.14B-Qwen2.5-Kunou-v1.Q4_K_M.gguf
+- !!merge <<: *qwen25
+  name: "dolphin3.0-qwen2.5-0.5b"
+  icon: https://cdn-uploads.huggingface.co/production/uploads/63111b2d88942700629f5771/cNCs1TBD3FelWCJGkZ3cd.png
+  urls:
+    - https://huggingface.co/cognitivecomputations/Dolphin3.0-Qwen2.5-0.5B
+    - https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-0.5B-GGUF
+  description: |
+    Dolphin 3.0 is the next generation of the Dolphin series of instruct-tuned models. Designed to be the ultimate general purpose local model, enabling coding, math, agentic, function calling, and general use cases.
+
+    Dolphin aims to be a general purpose model, similar to the models behind ChatGPT, Claude, Gemini. But these models present problems for businesses seeking to include AI in their products.
+
+        They maintain control of the system prompt, deprecating and changing things as they wish, often causing software to break.
+        They maintain control of the model versions, sometimes changing things silently, or deprecating older models that your business relies on.
+        They maintain control of the alignment, and in particular the alignment is one-size-fits all, not tailored to the application.
+        They can see all your queries and they can potentially use that data in ways you wouldn't want. Dolphin, in contrast, is steerable and gives control to the system owner. You set the system prompt. You decide the alignment. You have control of your data. Dolphin does not impose its ethics or guidelines on you. You are the one who decides the guidelines.
+
+    Dolphin belongs to YOU, it is your tool, an extension of your will. Just as you are personally responsible for what you do with a knife, gun, fire, car, or the internet, you are the creator and originator of any content you generate with Dolphin.
+  overrides:
+    parameters:
+      model: Dolphin3.0-Qwen2.5-0.5B-Q4_K_M.gguf
+  files:
+    - filename: Dolphin3.0-Qwen2.5-0.5B-Q4_K_M.gguf
+      sha256: 6a53689e2cb91027fdc9e366142eba8e35f56c14ee353e0a4d64de981efbfffa
+      uri: huggingface://bartowski/Dolphin3.0-Qwen2.5-0.5B-GGUF/Dolphin3.0-Qwen2.5-0.5B-Q4_K_M.gguf
+- !!merge <<: *qwen25
+  name: "dolphin3.0-qwen2.5-1.5b"
+  url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
+  icon: https://cdn-uploads.huggingface.co/production/uploads/63111b2d88942700629f5771/cNCs1TBD3FelWCJGkZ3cd.png
+  urls:
+    - https://huggingface.co/cognitivecomputations/Dolphin3.0-Qwen2.5-1.5B
+    - https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-1.5B-GGUF
+  description: |
+    Dolphin 3.0 is the next generation of the Dolphin series of instruct-tuned models. Designed to be the ultimate general purpose local model, enabling coding, math, agentic, function calling, and general use cases.
+
+    Dolphin aims to be a general purpose model, similar to the models behind ChatGPT, Claude, Gemini. But these models present problems for businesses seeking to include AI in their products.
+
+        They maintain control of the system prompt, deprecating and changing things as they wish, often causing software to break.
+        They maintain control of the model versions, sometimes changing things silently, or deprecating older models that your business relies on.
+        They maintain control of the alignment, and in particular the alignment is one-size-fits all, not tailored to the application.
+        They can see all your queries and they can potentially use that data in ways you wouldn't want. Dolphin, in contrast, is steerable and gives control to the system owner. You set the system prompt. You decide the alignment. You have control of your data. Dolphin does not impose its ethics or guidelines on you. You are the one who decides the guidelines.
+
+    Dolphin belongs to YOU, it is your tool, an extension of your will. Just as you are personally responsible for what you do with a knife, gun, fire, car, or the internet, you are the creator and originator of any content you generate with Dolphin.
+  overrides:
+    parameters:
+      model: Dolphin3.0-Qwen2.5-1.5B-Q4_K_M.gguf
+  files:
+    - filename: Dolphin3.0-Qwen2.5-1.5B-Q4_K_M.gguf
+      sha256: 7caa630a60c8831a509e2663e1761355fa24bcf6ccc03e3cc767e5b5747a3be5
+      uri: huggingface://bartowski/Dolphin3.0-Qwen2.5-1.5B-GGUF/Dolphin3.0-Qwen2.5-1.5B-Q4_K_M.gguf
+- !!merge <<: *qwen25
+  name: "dolphin3.0-qwen2.5-3b"
+  icon: https://cdn-uploads.huggingface.co/production/uploads/63111b2d88942700629f5771/cNCs1TBD3FelWCJGkZ3cd.png
+  url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
+  urls:
+    - https://huggingface.co/cognitivecomputations/Dolphin3.0-Qwen2.5-3b
+    - https://huggingface.co/bartowski/Dolphin3.0-Qwen2.5-3b-GGUF
+  description: |
+    Dolphin 3.0 is the next generation of the Dolphin series of instruct-tuned models. Designed to be the ultimate general purpose local model, enabling coding, math, agentic, function calling, and general use cases.
+
+    Dolphin aims to be a general purpose model, similar to the models behind ChatGPT, Claude, Gemini. But these models present problems for businesses seeking to include AI in their products.
+
+        They maintain control of the system prompt, deprecating and changing things as they wish, often causing software to break.
+        They maintain control of the model versions, sometimes changing things silently, or deprecating older models that your business relies on.
+        They maintain control of the alignment, and in particular the alignment is one-size-fits all, not tailored to the application.
+        They can see all your queries and they can potentially use that data in ways you wouldn't want. Dolphin, in contrast, is steerable and gives control to the system owner. You set the system prompt. You decide the alignment. You have control of your data. Dolphin does not impose its ethics or guidelines on you. You are the one who decides the guidelines.
+
+    Dolphin belongs to YOU, it is your tool, an extension of your will. Just as you are personally responsible for what you do with a knife, gun, fire, car, or the internet, you are the creator and originator of any content you generate with Dolphin.
+  overrides:
+    parameters:
+      model: Dolphin3.0-Qwen2.5-3b-Q4_K_M.gguf
+  files:
+    - filename: Dolphin3.0-Qwen2.5-3b-Q4_K_M.gguf
+      sha256: 0cb1908c5f444e1dc2c5b5619d62ac4957a22ad39cd42f2d0b48e2d8b1c358ab
+      uri: huggingface://bartowski/Dolphin3.0-Qwen2.5-3b-GGUF/Dolphin3.0-Qwen2.5-3b-Q4_K_M.gguf
 - &smollm
   ## SmolLM
   url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
@@ -4571,6 +4734,31 @@
     - filename: Llama3.1-8B-PRM-Deepseek-Data.Q4_K_M.gguf
       sha256: 254c7ccc4ea3818fe5f6e3ffd5500c779b02058b98f9ce9a3856e54106d008e3
       uri: huggingface://QuantFactory/Llama3.1-8B-PRM-Deepseek-Data-GGUF/Llama3.1-8B-PRM-Deepseek-Data.Q4_K_M.gguf
+- !!merge <<: *llama31
+  name: "dolphin3.0-llama3.1-8b"
+  url: "github:mudler/LocalAI/gallery/chatml.yaml@master"
+  icon: https://cdn-uploads.huggingface.co/production/uploads/63111b2d88942700629f5771/cNCs1TBD3FelWCJGkZ3cd.png
+  urls:
+    - https://huggingface.co/cognitivecomputations/Dolphin3.0-Llama3.1-8B
+    - https://huggingface.co/bartowski/Dolphin3.0-Llama3.1-8B-GGUF
+  description: |
+    Dolphin 3.0 is the next generation of the Dolphin series of instruct-tuned models. Designed to be the ultimate general purpose local model, enabling coding, math, agentic, function calling, and general use cases.
+
+    Dolphin aims to be a general purpose model, similar to the models behind ChatGPT, Claude, Gemini. But these models present problems for businesses seeking to include AI in their products.
+
+        They maintain control of the system prompt, deprecating and changing things as they wish, often causing software to break.
+        They maintain control of the model versions, sometimes changing things silently, or deprecating older models that your business relies on.
+        They maintain control of the alignment, and in particular the alignment is one-size-fits all, not tailored to the application.
+        They can see all your queries and they can potentially use that data in ways you wouldn't want. Dolphin, in contrast, is steerable and gives control to the system owner. You set the system prompt. You decide the alignment. You have control of your data. Dolphin does not impose its ethics or guidelines on you. You are the one who decides the guidelines.
+
+    Dolphin belongs to YOU, it is your tool, an extension of your will. Just as you are personally responsible for what you do with a knife, gun, fire, car, or the internet, you are the creator and originator of any content you generate with Dolphin.
+  overrides:
+    parameters:
+      model: Dolphin3.0-Llama3.1-8B-Q4_K_M.gguf
+  files:
+    - filename: Dolphin3.0-Llama3.1-8B-Q4_K_M.gguf
+      sha256: 268390e07edd407ad93ea21a868b7ae995b5950e01cad0db9e1802ae5049d405
+      uri: huggingface://bartowski/Dolphin3.0-Llama3.1-8B-GGUF/Dolphin3.0-Llama3.1-8B-Q4_K_M.gguf
 - &deepseek
   ## Deepseek
   url: "github:mudler/LocalAI/gallery/deepseek.yaml@master"
@@ -9289,6 +9477,29 @@
     - filename: Bio-Medical-Llama-3-8B.Q4_K_M.gguf
       sha256: 672939e0487d02c55734132c25a59f26e4deaac7cd49445a7028f2291139edcc
       uri: huggingface://QuantFactory/Bio-Medical-Llama-3-8B-GGUF/Bio-Medical-Llama-3-8B.Q4_K_M.gguf
+- !!merge <<: *llama3
+  name: "triangulum-10b"
+  icon: https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/By0OJ1lMvP5ZvVvfEGvz5.png
+  urls:
+    - https://huggingface.co/prithivMLmods/Triangulum-10B
+    - https://huggingface.co/mradermacher/Triangulum-10B-GGUF
+  description: |
+    Triangulum 10B is a collection of pretrained and instruction-tuned generative models, designed for multilingual applications. These models are trained using synthetic datasets based on long chains of thought, enabling them to perform complex reasoning tasks effectively.
+    Key Features
+        Foundation Model: Built upon LLaMA's autoregressive language model, leveraging an optimized transformer architecture for enhanced performance.
+        Instruction Tuning: Includes supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align model outputs with human preferences for helpfulness and safety.
+        Multilingual Support: Designed to handle multiple languages, ensuring broad applicability across diverse linguistic contexts.
+    Training Approach
+        Synthetic Datasets: Utilizes long chain-of-thought synthetic data to enhance reasoning capabilities.
+        Supervised Fine-Tuning (SFT): Aligns the model to specific tasks through curated datasets.
+        Reinforcement Learning with Human Feedback (RLHF): Ensures the model adheres to human values and safety guidelines through iterative training processes.
+  overrides:
+    parameters:
+      model: Triangulum-10B.Q4_K_M.gguf
+  files:
+    - filename: Triangulum-10B.Q4_K_M.gguf
+      sha256: dd071f99edf6b166044bf229cdeec19419c4c348e3fc3d6587cfcc55e6fb85fa
+      uri: huggingface://mradermacher/Triangulum-10B-GGUF/Triangulum-10B.Q4_K_M.gguf
 - &command-R
   ### START Command-r
   url: "github:mudler/LocalAI/gallery/command-r.yaml@master"