Paper Download Center

论文下载中心

所有 arXiv PDF 均已用 `HEAD` 请求验证可访问;MiniMax-M2.5 当前保留官方报告入口。

45可下载 PDF
92唯一外链已验证
2026-05-11最后校验日期

GLM Series

论文年份阅读页PDF
GLM: General Language Model Pre-training with Autoregressive Blank Infilling2021arXivPDF
GLM-130B: An Open Bilingual Pre-trained Model2022arXivPDF
WebGLM2023arXivPDF
ChatGLM / GLM-4 All Tools2024arXivPDF
AutoGLM: Autonomous Foundation Agents for GUIs2024arXivPDF
GLM-4-Voice2024arXivPDF
GLM-4.1V-Thinking & GLM-4.5V2025arXivPDF
GLM-4.5: Agentic, Reasoning, and Coding Foundation Models2025arXivPDF
GLM-TTS Technical Report2025arXivPDF
GLM-5: From Vibe Coding to Agentic Engineering2026arXivPDF
GLM-OCR Technical Report2026arXivPDF
GLM-5V-Turbo2026arXivPDF

Kimi Series

论文年份阅读页PDF
Mooncake: A KVCache-Centric Disaggregated Architecture for LLM Serving2024arXivPDF
Kimi k1.5: Scaling Reinforcement Learning with LLMs2025arXivPDF
Muon is Scalable for LLM Training2025arXivPDF
Kimi-VL Technical Report2025arXivPDF
Kimi-Audio Technical Report2025arXivPDF
Kimi K2: Open Agentic Intelligence2025/2026arXivPDF
Kimi Linear: An Expressive, Efficient Attention Architecture2025arXivPDF
Kimi K2.5: Visual Agentic Intelligence2026arXivPDF

DeepSeek Series

论文年份阅读页PDF
DeepSeek LLM: Scaling Open-Source Language Models with Longtermism2024arXivPDF
DeepSeekMoE: Towards Ultimate Expert Specialisation in Mixture-of-Experts Language Models2024arXivPDF
DeepSeek-Coder: When the Large Language Model Meets Programming2024arXivPDF
DeepSeek-Math: Pushing the Limits of Mathematical Reasoning2024arXivPDF
DeepSeek-VL: Towards Real-World Vision-Language Understanding2024arXivPDF
DeepSeek-V2: A Strong, Economical and Efficient Mixture-of-Experts Language Model2024arXivPDF
DeepSeek-Prover2024arXivPDF
DeepSeek-Coder-V22024arXivPDF
Let the Expert Stick to His Last: Expert-Specialised Fine-Tuning for Sparse Models2024arXivPDF
Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts2024arXivPDF
DeepSeek-Prover V1.52024arXivPDF
Janus2024arXivPDF
JanusFlow2024arXivPDF
DeepSeek-VL22024arXivPDF
DeepSeek-V3 Technical Report2024arXivPDF
Janus-Pro2025arXivPDF
DeepSeek-R12025arXivPDF
Native Sparse Attention2025arXivPDF
Inference-Time Scaling for Generalist Reward Modelling2025arXivPDF
DeepSeek-Prover V22025arXivPDF
DeepSeek-OCR2025arXivPDF
DeepSeek-Math-V22025arXivPDF
DeepSeek-V3.22025arXivPDF

MiniMax Series

论文/报告年份阅读页PDF / 报告
MiniMax-01: Scaling Foundation Models with Lightning Attention2025arXivPDF
MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention2025arXivPDF
MiniMax M2.5: Built for Real-World Productivity2026OfficialModel page