SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2412.03555 · 2024

PaliGemma 2: A Family of Versatile VLMs for Transfer

arXiv · PDF · Open in the Atlas

Code that ran

4 of 7 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, 0 violated it and 4 have no contract verdict; 0 failures of our instrument; 3 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/2412.03555.json · badge: /api/ran/2412.03555.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
kyutai-labs/moshivis community (classifier label)
classifier basis not recorded on the link
4 ran · 3 no recorded run
FunctionStatusWhere it lives
colorize Ran
ran on a synthesized input
kyutai-labs/moshivis/kyuteye_mlx/kyuteye_mlx/local_web.py
code served (permissive licence) · get_code("a11e8571b42d5bf8")
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 str. 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

# Analyze the function:
# - text: str - needs a string value
# - color: str - needs a string value (ANSI color code like '32' for green, '31' for red, etc.)

ARGS = {
    'text': 'Hello World',
    'color': '32'
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
compile_pattern Ran
ran on a synthesized input
kyutai-labs/moshivis/ssvd/utils.py
code served (permissive licence) · get_code("fcb8bc1bcc0e936a")
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 Pattern. 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 re
from functools import lru_cache
from typing import Dict, List, Pattern, Sequence, Tuple

@lru_cache
def compile_pattern(s: str) -> Pattern:
    """cached compile"""
    return re.compile(s)

# Setup code to construct valid arguments
ARGS = {
    "s": r"^\d+$"
}
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 Pattern — a Pattern output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Pattern 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
maybe_shorten_caption Ran
ran on a synthesized input
kyutai-labs/moshivis/ssvd/utils.py
code served (permissive licence) · get_code("250e3cffe0703509")
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 str. 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

# Test case 1: caption shorter than max length
caption_short = "This is a short caption."
max_cap_len_short = 1500

# Test case 2: caption longer than max length with multiple sentences
caption_long = "This is the first sentence. This is the second sentence. This is the third sentence. This is the fourth sentence."
max_cap_len_long = 50

# Test case 3: caption with edge case of double periods
caption_double_period = "This is a sentence.. This is another sentence."
max_cap_len_double = 1500

# Test case 4: very long caption that needs truncation to exact character limit
caption_very_long = "A" * 2000
max_cap_len_very_long = 100

# Use the first test case for ARGS
ARGS = {
    "caption": caption_short,
    "max_cap_len": max_cap_len_short
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
preprocess_pixelprose_captions Ran
ran on a synthesized input
kyutai-labs/moshivis/ssvd/utils.py
code served (permissive licence) · get_code("e6c9ef475f43f2f7")
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 dict. 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)

# Analyze the function signature
# preprocess_pixelprose_captions(caption: str) -> Dict[str, str]
# The function takes a single parameter: caption (str)

# Test with a valid caption that exercises the function logic
caption = "This image displays: The image is a beautiful landscape. This image is taken in the mountains. Some other sentence. The background is blue."

ARGS = {
    'caption': caption
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
get_model_file Not yet run
unverified (harvested, no recorded run)
kyutai-labs/moshivis/kyuteye_mlx/kyuteye_mlx/local_web.py
code served (permissive licence) · get_code("e65b1f630e98892d")
How it ran

no invocation recorded.

get_pipeline Not yet run
unverified (harvested, no recorded run)
kyutai-labs/moshivis/ssvd/generate.py
code served (permissive licence) · get_code("b75902c10eb95a50")
How it ran

no invocation recorded.

hf_hub_download Not yet run
unverified (harvested, no recorded run)
kyutai-labs/moshivis/kyuteye_mlx/kyuteye_mlx/local_web.py
code served (permissive licence) · get_code("d98ebfa2e7a5c6d3")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 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.

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

PaliGemma 2 is an upgrade of the PaliGemma open Vision-Language Model (VLM) based on the Gemma 2 family of language models. We combine the SigLIP-So400m vision encoder that was also used by PaliGemma with the whole range of Gemma 2 models, from the 2B one all the way up to the 27B model. We train these models at three resolutions (224px, 448px, and 896px) in multiple stages to equip them with broad knowledge for transfer via fine-tuning. The resulting family of base models covering different model sizes and resolutions allows us to investigate factors impacting transfer performance (such as learning rate) and to analyze the interplay between the type of task, model size, and resolution. We further increase the number and breadth of transfer tasks beyond the scope of PaliGemma including different OCR-related tasks such as table structure recognition, molecular structure recognition, music score recognition, as well as long fine-grained captioning and radiography report generation, on which PaliGemma 2 obtains state-of-the-art results.

For agents

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

get_harvested_code_for_paper("2412.03555")
get_code_for_paper("2412.03555")
have("2412.03555")

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

curl https://syntology.ai/api/ran/2412.03555.json

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

[![Syntology run record](https://syntology.ai/api/ran/2412.03555.svg)](https://syntology.ai/paper/2412.03555)

Connect an agent — have() is free.