server: add webp image input support (#10653)

cmd/interactive.go

@@ -44,7 +44,7 @@ func generateInteractive(cmd *cobra.Command, opts runOptions) error {
 		fmt.Fprintln(os.Stderr, "Use \"\"\" to begin a multi-line message.")
 
 		if opts.MultiModal {
-			fmt.Fprintf(os.Stderr, "Use %s to include .jpg or .png images.\n", filepath.FromSlash("/path/to/file"))
+			fmt.Fprintf(os.Stderr, "Use %s to include .jpg, .png, or .webp images.\n", filepath.FromSlash("/path/to/file"))
 		}
 
 		fmt.Fprintln(os.Stderr, "")
@@ -511,7 +511,7 @@ func extractFileNames(input string) []string {
 	// Regex to match file paths starting with optional drive letter, / ./ \ or .\ and include escaped or unescaped spaces (\ or %20)
 	// and followed by more characters and a file extension
 	// This will capture non filename strings, but we'll check for file existence to remove mismatches
-	regexPattern := `(?:[a-zA-Z]:)?(?:\./|/|\\)[\S\\ ]+?\.(?i:jpg|jpeg|png)\b`
+	regexPattern := `(?:[a-zA-Z]:)?(?:\./|/|\\)[\S\\ ]+?\.(?i:jpg|jpeg|png|webp)\b`
 	re := regexp.MustCompile(regexPattern)
 
 	return re.FindAllString(input, -1)
@@ -553,7 +553,7 @@ func getImageData(filePath string) ([]byte, error) {
 	}
 
 	contentType := http.DetectContentType(buf)
-	allowedTypes := []string{"image/jpeg", "image/jpg", "image/png"}
+	allowedTypes := []string{"image/jpeg", "image/jpg", "image/png", "image/webp"}
 	if !slices.Contains(allowedTypes, contentType) {
 		return nil, fmt.Errorf("invalid image type: %s", contentType)
 	}

cmd/interactive_test.go

@@ -12,14 +12,17 @@ func TestExtractFilenames(t *testing.T) {
 	// Unix style paths
 	input := ` some preamble 
  ./relative\ path/one.png inbetween1 ./not a valid two.jpg inbetween2 ./1.svg
-/unescaped space /three.jpeg inbetween3 /valid\ path/dir/four.png "./quoted with spaces/five.JPG`
+/unescaped space /three.jpeg inbetween3 /valid\ path/dir/four.png "./quoted with spaces/five.JPG
+/unescaped space /six.webp inbetween6 /valid\ path/dir/seven.WEBP`
 	res := extractFileNames(input)
-	assert.Len(t, res, 5)
+	assert.Len(t, res, 7)
 	assert.Contains(t, res[0], "one.png")
 	assert.Contains(t, res[1], "two.jpg")
 	assert.Contains(t, res[2], "three.jpeg")
 	assert.Contains(t, res[3], "four.png")
 	assert.Contains(t, res[4], "five.JPG")
+	assert.Contains(t, res[5], "six.webp")
+	assert.Contains(t, res[6], "seven.WEBP")
 	assert.NotContains(t, res[4], '"')
 	assert.NotContains(t, res, "inbetween1")
 	assert.NotContains(t, res, "./1.svg")
@@ -30,10 +33,12 @@ func TestExtractFilenames(t *testing.T) {
  /absolute/nospace/three.jpeg inbetween3 /absolute/with space/four.png inbetween4
 ./relative\ path/five.JPG inbetween5 "./relative with/spaces/six.png inbetween6
 d:\path with\spaces\seven.JPEG inbetween7 c:\users\jdoe\eight.png inbetween8 
- d:\program files\someplace\nine.png inbetween9 "E:\program files\someplace\ten.PNG some ending
+ d:\program files\someplace\nine.png inbetween9 "E:\program files\someplace\ten.PNG
+c:/users/jdoe/eleven.webp inbetween11 c:/program files/someplace/twelve.WebP inbetween12
+d:\path with\spaces\thirteen.WEBP some ending
 `
 	res = extractFileNames(input)
-	assert.Len(t, res, 10)
+	assert.Len(t, res, 13)
 	assert.NotContains(t, res, "inbetween2")
 	assert.Contains(t, res[0], "one.png")
 	assert.Contains(t, res[0], "c:")
@@ -51,6 +56,12 @@ d:\path with\spaces\seven.JPEG inbetween7 c:\users\jdoe\eight.png inbetween8
 	assert.Contains(t, res[8], "d:")
 	assert.Contains(t, res[9], "ten.PNG")
 	assert.Contains(t, res[9], "E:")
+	assert.Contains(t, res[10], "eleven.webp")
+	assert.Contains(t, res[10], "c:")
+	assert.Contains(t, res[11], "twelve.WebP")
+	assert.Contains(t, res[11], "c:")
+	assert.Contains(t, res[12], "thirteen.WEBP")
+	assert.Contains(t, res[12], "d:")
 }
 
 // Ensure that file paths wrapped in single quotes are removed with the quotes.

server/routes.go

@@ -8,6 +8,7 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"image"
 	"io"
 	"io/fs"
 	"log/slog"
@@ -25,6 +26,7 @@ import (
 
 	"github.com/gin-contrib/cors"
 	"github.com/gin-gonic/gin"
+	"golang.org/x/image/webp"
 	"golang.org/x/sync/errgroup"
 
 	"github.com/ollama/ollama/api"
@@ -1304,6 +1306,10 @@ func Serve(ln net.Listener) error {
 
 	s.sched.Run(schedCtx)
 
+	// register the experimental webp decoder
+	// so webp images can be used in multimodal inputs
+	image.RegisterFormat("webp", "RIFF????WEBP", webp.Decode, webp.DecodeConfig)
+
 	// At startup we retrieve GPU information so we can get log messages before loading a model
 	// This will log warnings to the log in case we have problems with detected GPUs
 	gpus := discover.GetGPUInfo()