Merge pull request #14 from yethee/gpt4o

Added support for GPT-4o
yethee · May 14, 2024 · b6348d7 · b6348d7
2 parents f512721 + ab36387
commit b6348d7
Show file tree

Hide file tree

Showing 4 changed files with 200,032 additions and 11 deletions.
diff --git a/src/EncoderProvider.php b/src/EncoderProvider.php
@@ -36,14 +36,21 @@ final class EncoderProvider implements ResetInterface
             'vocab' => 'https://openaipublic.blob.core.windows.net/encodings/cl100k_base.tiktoken',
             'pat' => '/(?i:\'s|\'t|\'re|\'ve|\'m|\'ll|\'d)|[^\r\n\p{L}\p{N}]?\p{L}+|\p{N}{1,3}| ?[^\s\p{L}\p{N}]+[\r\n]*|\s*[\r\n]+|\s+(?!\S)|\s+/u',
         ],
+        'o200k_base' => [
+            'vocab' => 'https://openaipublic.blob.core.windows.net/encodings/o200k_base.tiktoken',
+            'pat' => '/[^\r\n\p{L}\p{N}]?[\p{Lu}\p{Lt}\p{Lm}\p{Lo}\p{M}]*[\p{Ll}\p{Lm}\p{Lo}\p{M}]+(?i:\'s|\'t|\'re|\'ve|\'m|\'ll|\'d)?|[^\r\n\p{L}\p{N}]?[\p{Lu}\p{Lt}\p{Lm}\p{Lo}\p{M}]+[\p{Ll}\p{Lm}\p{Lo}\p{M}]*(?i:\'s|\'t|\'re|\'ve|\'m|\'ll|\'d)?|\p{N}{1,3}| ?[^\s\p{L}\p{N}]+[\r\n\/]*|\s*[\r\n]+|\s+(?!\S)|\s+/u',
+        ],
     ];
     private const MODEL_PREFIX_TO_ENCODING = [
+        'gpt-4o-' => 'o200k_base',
         'gpt-4-' => 'cl100k_base',
         'gpt-3.5-turbo-' => 'cl100k_base',
     ];
     private const MODEL_TO_ENCODING = [
+        'gpt-4o' => 'o200k_base',
         'gpt-4' => 'cl100k_base',
         'gpt-3.5-turbo' => 'cl100k_base',
+        'gpt-3.5' => 'cl100k_base',
         'text-davinci-003' => 'p50k_base',
         'text-davinci-002' => 'p50k_base',
         'text-davinci-001' => 'r50k_base',

diff --git a/tests/EncoderProviderTest.php b/tests/EncoderProviderTest.php
@@ -49,5 +49,7 @@ public static function getEncoderForModelProvider(): iterable
         yield 'text-davinci-003' => ['text-davinci-003', 'p50k_base'];
         yield 'text-davinci-edit-001' => ['text-davinci-edit-001', 'p50k_edit'];
         yield 'gpt-3.5-turbo-0301' => ['gpt-3.5-turbo-0301', 'cl100k_base'];
+        yield 'gpt-4-32k' => ['gpt-4-32k', 'cl100k_base'];
+        yield 'gpt-4o-2024-05-13' => ['gpt-4o-2024-05-13', 'o200k_base'];
     }
 }
diff --git a/tests/EncoderTest.php b/tests/EncoderTest.php
@@ -12,20 +12,26 @@
 
 final class EncoderTest extends TestCase
 {
-    /** @param list<int> $tokens */
+    /**
+     * @param non-empty-string $encoding
+     * @param list<int>        $tokens
+     */
     #[DataProvider('provideDataForFlatTokenization')]
-    public function testEncode(string $text, array $tokens): void
+    public function testEncode(string $text, string $encoding, array $tokens): void
     {
-        $encoder = self::getEncoder('cl100k_base');
+        $encoder = self::getEncoder($encoding);
 
         self::assertSame($tokens, $encoder->encode($text));
     }
 
-    /** @param list<int> $tokens */
+    /**
+     * @param non-empty-string $encoding
+     * @param list<int>        $tokens
+     */
     #[DataProvider('provideDataForFlatTokenization')]
-    public function testDecode(string $text, array $tokens): void
+    public function testDecode(string $text, string $encoding, array $tokens): void
     {
-        $encoder = self::getEncoder('cl100k_base');
+        $encoder = self::getEncoder($encoding);
 
         self::assertSame($text, $encoder->decode($tokens));
     }
@@ -41,19 +47,27 @@ public function testEncodeInChunks(Encoder $encoder, string $text, int $maxToken
     }
 
     /**
-     * @return iterable<array{string, list<int>}>
+     * @return iterable<array{string, string, list<int>}>
      *
      * @psalm-suppress PossiblyUnusedMethod
      */
     public static function provideDataForFlatTokenization(): iterable
     {
-        yield 'hello world' => ['hello world', [15339, 1917]];
+        yield '[cl100k_base] hello world' => ['hello world', 'cl100k_base', [15339, 1917]];
+
+        yield '[cl100k_base] привет мир' => ['привет мир', 'cl100k_base', [8164, 2233, 28089, 8341, 11562, 78746]];
+
+        yield '[cl100k_base] emoji' => ['🌶', 'cl100k_base', [9468, 234, 114]];
+
+        yield '[cl100k_base] new line character' => [".\n", 'cl100k_base', [627]];
+
+        yield '[o200k_base] hello world' => ['hello world', 'o200k_base', [24912, 2375]];
 
-        yield 'привет мир' => ['привет мир', [8164, 2233, 28089, 8341, 11562, 78746]];
+        yield '[o200k_base] привет мир' => ['привет мир', 'o200k_base', [9501, 131903, 37934]];
 
-        yield 'emoji' => ['🌶', [9468, 234, 114]];
+        yield '[o200k_base] emoji' => ['🌶', 'o200k_base', [64364, 114]];
 
-        yield 'new line character' => [".\n", [627]];
+        yield '[o200k_base] new line character' => [".\n", 'o200k_base', [558]];
     }
 
     /**