From 52acf840a6b15e86a507f55c26662a2301cafcef Mon Sep 17 00:00:00 2001
From: suh016 <afifsabrin25@gmail.com>
Date: Mon, 4 May 2026 17:11:42 +0800
Subject: [PATCH] Allow callers to pass tokenizer_options in
 Text2TextGenerationPipeline

The tokenizer call inside _call() previously used hardcoded options with
no way to override them. Now callers can pass tokenizer_options alongside
generation kwargs to control things like max_length and truncation_side.

tokenizer_options is extracted from generate_kwargs before model.generate()
is called so it never leaks into GenerationFunctionParameters. Pipeline
defaults (padding: true, truncation: true) still apply and caller values
take precedence.

Closes #1096.
---
 .../src/pipelines/text2text-generation.js      | 11 +++++++++--
 .../test_pipelines_text2text_generation.js     | 18 ++++++++++++++++++
 2 files changed, 27 insertions(+), 2 deletions(-)

diff --git a/packages/transformers/src/pipelines/text2text-generation.js b/packages/transformers/src/pipelines/text2text-generation.js
index b313dd91f..700c59552 100644
--- a/packages/transformers/src/pipelines/text2text-generation.js
+++ b/packages/transformers/src/pipelines/text2text-generation.js
@@ -72,16 +72,23 @@ export class Text2TextGenerationPipeline
         }
 
         const tokenizer = this.tokenizer;
+
+        // Callers may pass tokenizer_options as a top-level key inside generate_kwargs to
+        // control tokenization (e.g. max_length, truncation side) without touching generation
+        // parameters.  We extract it here so it never leaks into model.generate().
+        const { tokenizer_options: caller_tokenizer_options, ...rest_generate_kwargs } = generate_kwargs;
         const tokenizer_options = {
             padding: true,
             truncation: true,
+            ...caller_tokenizer_options,
         };
+
         let inputs;
         if (this.task === 'translation' && '_build_translation_inputs' in tokenizer) {
             // TODO: move to Translation pipeline?
             // Currently put here to avoid code duplication
             // @ts-ignore
-            inputs = tokenizer._build_translation_inputs(texts, tokenizer_options, generate_kwargs);
+            inputs = tokenizer._build_translation_inputs(texts, tokenizer_options, rest_generate_kwargs);
         } else {
             inputs = tokenizer(texts, tokenizer_options);
         }
@@ -89,7 +96,7 @@ export class Text2TextGenerationPipeline
         const outputTokenIds = await this.model.generate({
             ...inputs,
             ...this._default_generation_config,
-            ...generate_kwargs,
+            ...rest_generate_kwargs,
         });
         return tokenizer
             .batch_decode(/** @type {Tensor} */ (outputTokenIds), {
diff --git a/packages/transformers/tests/pipelines/test_pipelines_text2text_generation.js b/packages/transformers/tests/pipelines/test_pipelines_text2text_generation.js
index d8d756f34..3abd4e033 100644
--- a/packages/transformers/tests/pipelines/test_pipelines_text2text_generation.js
+++ b/packages/transformers/tests/pipelines/test_pipelines_text2text_generation.js
@@ -31,6 +31,24 @@ export default () => {
         },
         MAX_TEST_EXECUTION_TIME,
       );
+
+      it(
+        "tokenizer_options are forwarded to the tokenizer",
+        async () => {
+          const text = "This is a test.";
+          // max_length=3 forces aggressive truncation via tokenizer_options;
+          // the model should still run without error and return a string result.
+          // tokenizer_options is extracted before generate() is called so it
+          // never leaks into GenerationFunctionParameters.
+          const output = await pipe(text, {
+            max_new_tokens: 5,
+            tokenizer_options: { truncation: true, max_length: 3 },
+          });
+          expect(Array.isArray(output)).toBe(true);
+          expect(typeof output[0].generated_text).toBe("string");
+        },
+        MAX_TEST_EXECUTION_TIME,
+      );
     });
 
     afterAll(async () => {