SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.05901 · 2025

Valley2: Exploring Multimodal Models with Scalable Vision-Language Design

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 6 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2501.05901.json · badge: /api/ran/2501.05901.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
bytedance/valley canonical (classifier label)
classifier basis not recorded on the link
7 ran · 1 no recorded run
FunctionStatusWhere it lives
b64_decode Ran
ran on a synthesized input
bytedance/valley/valley/util/decode_img.py
code served (permissive licence) · get_code("8519d8e6d16f76d6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned bytes. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import base64

# Create a valid base64-encoded string
original_string = "Hello, World!"
encoded_string = base64.b64encode(original_string.encode()).decode()

ARGS = {
    "string": encoded_string
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bytes — a bytes output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a bytes output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
check_model_config Ran
ran on a synthesized input
bytedance/valley/valley2/valley_utils.py
code served (permissive licence) · get_code("5826cc219ebd0f66")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Config. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import warnings

class Config:
    def __init__(self, model_class=None, llm_name=None):
        self.model_class = model_class
        self.llm_name = llm_name

# Test case 1: Valid configuration with valley-product and llama
config = Config(model_class='valley-product', llm_name='llama')

ARGS = {
    'config': config
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Config — a Config output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Config output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
collate_process_image_text Ran
ran on a synthesized input
bytedance/valley/valley/util/data_util.py
code served (permissive licence) · get_code("e1d6c2a78ced9904")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

class MockTokenizer:
    def __call__(self, text, padding=True):
        # Return a mock tokenizer output with input_ids and attention_mask
        class TokenizerOutput:
            def __init__(self):
                self.input_ids = [[101, 2054, 0], [101, 2009, 0]]
                self.attention_mask = [[1, 1, 0], [1, 1, 0]]
        return TokenizerOutput()

class MockImageProcessor:
    def preprocess(self, images, return_tensors=None):
        # Return a dict with pixel_values tensor
        # Simulate processing a batch of images
        pixel_values = torch.randn(1, 3, 4, 4)
        return {'pixel_values': pixel_values}

class MockBatch:
    def __iter__(self):
        # Return batch as tuple: (batch_prompt, batch_image, conv_list, label_list)
        batch_prompt = ["What is this?", "Describe this"]
        # Each element in batch_image is a list of images for that sample
        batch_image = [
            [torch.randn(3, 8, 8), torch.randn(3, 8, 8)],
            [torch.randn(3, 8, 8)]
        ]
        conv_list = ["conv1", "conv2"]
        label_list = [0, 1]
        return iter((batch_prompt, batch_image, conv_list, label_list))

tokenizer = MockTokenizer()
image_processor = MockImageProcessor()
batch = MockBatch()

ARGS = {
    "batch": batch,
    "tokenizer": tokenizer,
    "image_processor": image_processor
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
collate_wrapper Ran
ran on a synthesized input
bytedance/valley/valley/util/data_util.py
code served (permissive licence) · get_code("40f75b6a1d7442f3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The collate_wrapper function expects a batch where each element b is a tuple/list with:
# b[0] = image
# b[1] = label
# b[2] = prompt
# b[3] = conv

# Create sample batch items
batch_item_1 = (
    torch.randn(4, 3),  # b[0]: image tensor (4x3)
    "label_1",           # b[1]: label
    "prompt_1",          # b[2]: prompt
    "conv_1"             # b[3]: conv
)

batch_item_2 = (
    torch.randn(4, 3),  # b[0]: image tensor (4x3)
    "label_2",           # b[1]: label
    "prompt_2",          # b[2]: prompt
    "conv_2"             # b[3]: conv
)

batch = [batch_item_1, batch_item_2]

ARGS = {
    "batch": batch
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_logger Ran
ran on a synthesized input
bytedance/valley/valley/utils.py
code served (permissive licence) · get_code("bc264e58e84dcfbe")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Logger. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import logging
import sys

ARGS = {
    "name": "test_logger"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Logger — a Logger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Logger output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
image_preprocess Ran
ran on a synthesized input
bytedance/valley/valley/util/decode_img.py
code served (permissive licence) · get_code("54802dd67899555a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Image. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import base64
from PIL import Image
import io

# Create a small test image (RGB, 10x10 pixels)
test_image = Image.new('RGB', (10, 10), color='red')

# Convert image to bytes
img_buffer = io.BytesIO()
test_image.save(img_buffer, format='PNG')
img_bytes = img_buffer.getvalue()

# Encode to base64 string
image_b64_str = base64.b64encode(img_bytes).decode('utf-8')

ARGS = {
    'image_str': image_b64_str
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
violates_moderation Ran
ran, violated the contract we drafted
bytedance/valley/valley/utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Mi3-Lab/doScenes-VLM-Planning, is pointer only (licence: AGPL-3.0) · get_code("f9939a84b9a65279")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from Mi3-Lab/doScenes-VLM-Planning, and AGPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

preprocess_multimodal Not yet run
unverified (harvested, no recorded run)
bytedance/valley/valley/util/data_util.py
code served (permissive licence) · get_code("bdb4a5432865c532")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
violates_moderation 56 other papers
48 dated earlier, 6 later, 2 same year, order not determinable
2106.06909 (2021-06)
GigaSpeech: An Evolving, Multi-domain ASR Corpus with 10,000 Hours of Transcribed Audio
2308.03279 (2023-08)
UniversalNER: TARGETED DISTILLATION FROM LARGE LANGUAGE MODELS FOR OPEN NAMED ENTITY RECOGNITION
2308.13437 (2023-08)
Position-Enhanced Visual Instruction Tuning for Multimodal Large Language Models
and 53 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Recently, vision-language models have made remarkable progress, demonstrating outstanding capabilities in various tasks such as image captioning and video understanding. We introduce Valley2, a novel multimodal large language model designed to enhance performance across all domains and extend the boundaries of practical applications in e-commerce and short video scenarios. Notably, Valley2 achieves state-of-the-art (SOTA) performance on e-commerce benchmarks, surpassing open-source models of similar size by a large margin (79.66 vs. 72.76). Additionally, Valley2 ranks second on the OpenCompass leaderboard among models with fewer than 10B parameters, with an impressive average score of 67.4. The code and model weights are open-sourced at https://github.com/bytedance/Valley.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2501.05901")
get_code_for_paper("2501.05901")
have("2501.05901")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2501.05901.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2501.05901.svg)](https://syntology.ai/paper/2501.05901)

Connect an agent — have() is free.