SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.08313 · 2025

MiniMax-01: Scaling Foundation Models with Lightning Attention

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 2 failures of our instrument; 0 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/2501.08313.json · badge: /api/ran/2501.08313.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
MiniMax-AI/MiniMax-01 canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
FunctionStatusWhere it lives
modify_gt Ran
ran; our contract draft was wrong, not the code
MiniMax-AI/MiniMax-01/evaluation/MR-NIAH/score.py
code served (permissive licence) · get_code("44e5fac8cbac336d")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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.

# Setup code for modify_gt function

# The function takes a single parameter 'gt' which is a string
# It matches against various predefined strings and returns a list of strings

# Valid test cases from the function's match statement
valid_gt_values = [
    '1. 钢琴\n2. 小提琴\n3. 吉他',
    '1. 生机勃勃\n2. 春暖花开\n3. 万物复苏',
    '体型小巧,羽毛灰褐,\n喜欢在城市中觅食,叽叽喳喳很热闹。',
    '1. 韩信\n2. 岳飞\n3. 霍去病',
    '蔚蓝无垠,波涛汹涌,生命的摇篮。',
    '1. 苹果\n2. 香蕉\n3. 橙子',
    '蝉鸣阵阵,知了此起彼伏。\n树荫下,老人们悠闲地下着棋。\n孩童嬉戏,欢笑声传遍公园。',
    '1. 微积分\n2. 线性代数\n3. 概率论',
    '红艳如火,娇嫩欲滴,\n花瓣层叠,芳香四溢。',
    '在南极的冰山之巅,\n企鹅们舞动着短小的翅膀。\n身披黑白礼服,步伐蹒跚,\n在寒风中,它们笑对严霜。',
    'On the peak of the Antarctic iceberg,\nPenguins dance with tiny wings.\nWearing black and white tuxedos, stumbling steps,\nThey smile at the severe frost in the cold wind.',
    'Red as fire, delicate and dripping,\nPetals layered, fragrance overflowing.',
    '1. Calculus\n2. Linear Algebra\n3. Probability Theory',
    'Cicadas chirping, the sounds rise and fall.\nUnder the shade, elders leisurely play chess.\nChildren play, laughter fills the park.',
    '1. Apple\n2. Banana\n3. Orange',
    'Vast and blue, waves surging, cradle of life.',
    '1. Han Xin\n2. Yue Fei\n3. Huo Qubing',
    'Small in size, gray-brown feathers,\nLikes to forage in the city, chirping lively.',
    '1. Piano\n2. Violin\n3. Guitar',
    '1. Vibrant\n2. Fresh\n3. Warm',
]

# Use one of the valid values as the argument
ARGS = {
    'gt': '1. 钢琴\n2. 小提琴\n3. 吉他'
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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
process_log Ran
ran; our contract draft was wrong, not the code
MiniMax-AI/MiniMax-01/evaluation/MR-NIAH/score.py
code served (permissive licence) · get_code("f53a39974983bad3")
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, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 function expects response_log to be a dict with at least a 'label' key
response_log = {
    'label': 'positive',
    'other_data': 'some_value'
}

ARGS = {
    'response_log': response_log
}
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, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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
score_response Ran
ran, honoured the contract we drafted
MiniMax-AI/MiniMax-01/evaluation/MR-NIAH/score.py
code served (permissive licence) · get_code("4e54cd947d374327")
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 float64, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 numpy as np

ARGS = {
    'response': 'The piano is a beautiful instrument. The violin produces wonderful sounds. The guitar is versatile and popular.',
    'gt_label': '1. 钢琴\n2. 小提琴\n3. 吉他',
    'language': 'chinese'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 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 3 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

We introduce MiniMax-01 series, including MiniMax-Text-01 and MiniMax-VL-01, which are comparable to top-tier models while offering superior capabilities in processing longer contexts. The core lies in lightning attention and its efficient scaling. To maximize computational capacity, we integrate it with Mixture of Experts (MoE), creating a model with 32 experts and 456 billion total parameters, of which 45.9 billion are activated for each token. We develop an optimized parallel strategy and highly efficient computation-communication overlap techniques for MoE and lightning attention. This approach enables us to conduct efficient training and inference on models with hundreds of billions of parameters across contexts spanning millions of tokens. The context window of MiniMax-Text-01 can reach up to 1 million tokens during training and extrapolate to 4 million tokens during inference at an affordable cost. Our vision-language model, MiniMax-VL-01 is built through continued training with 512 billion vision-language tokens. Experiments on both standard and in-house benchmarks show that our models match the performance of state-of-the-art models like GPT-4o and Claude-3.5-Sonnet while offering 20-32 times longer context window. We publicly release MiniMax-01 at https://github.com/MiniMax-AI.

For agents

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

get_harvested_code_for_paper("2501.08313")
get_code_for_paper("2501.08313")
have("2501.08313")

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

curl https://syntology.ai/api/ran/2501.08313.json

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

[![Syntology run record](https://syntology.ai/api/ran/2501.08313.svg)](https://syntology.ai/paper/2501.08313)

Connect an agent — have() is free.