Introducing Kimi K3.1 ProWe're introducing Kimi K3.1 Pro , the first model in our new K3.1 series and our most capable 3T-class open-weight model to date, with a 1-million-token context window. It delivers substantial improvements across coding, 3D creation, financKimi10/24 00:00
Introducing PerceptionBenchEvaluating Atomic Visual Perception in Multimodal Large Language Models Authors Kimi Team Overview We are releasing PerceptionBench , a benchmark that isolates visual perception and evaluates it as a set of atomic capabilities— discovered fKimi07/16 00:00
Kimi K3: Open Frontier IntelligenceToday, we are introducing Kimi K3 — our most capable model. Kimi K3 is a 2.8T-parameter model built on our Kimi Delta Attention and Attention Residuals, with native vision capabilities and a 1-million-token context window. It is the world'sKimi07/16 00:00
Kimi K2.6: Advancing Open-Source CodingWe are open sourcing our latest model, Kimi K2.6 , featuring state-of-the-art coding, long-horizon execution, and agent swarm capabilities . Kimi K2.6 is now available via Kimi.com , the Kimi App, the API , and Kimi Code . Benchmark Kimi K3Kimi04/20 00:00
Kimi Introduces Agent Swarm: Let 100 AI Agents Work for YouIn 2025, if you walked into any AI conference, you may hear the same gospel: faster inference, longer context windows, cheaper inference costs. It's as if we've spent years perfecting the hammer, making it lighter, stronger, more precisely Kimi02/09 00:00
Introducing WorldVQAA benchmark for evaluating atomic visual world knowledge in Multimodal LLMs. Authors Kimi Team Overview We are releasing WorldVQA , a new benchmark designed to measure the factual correctness of Multimodal Large Language Models (MLLMs). WhiKimi02/03 00:00
Kimi K2.5: Visual Agentic IntelligenceToday, we are introducing Kimi K2.5, the most powerful open-source model to date. Kimi K2.5 builds on Kimi K2 with continued pretraining over approximately 15T mixed visual and text tokens. Built as a native multimodal model, K2.5 delivers Kimi01/27 00:00
Rebuilding the "Chain of Trust": Kimi Vendor VerifierAlongside the release of the Kimi K2.6 model, we are open-sourcing the Kimi Vendor Verifier (KVV) project, designed to help users of open-source models verify the accuracy of their inference implementations. Not as an afterthought, but becaKimi01/22 00:00
Introducing Kimi K2 ThinkingToday, we are introducing Kimi K2 Thinking , our best open-source thinking model. Built as a thinking agent , it reasons step by step while using tools, achieving state-of-the-art performance on Humanity's Last Exam (HLE), BrowseComp, and oKimi11/06 00:00
Kimi-K2-Instruct-0905📰 Tech Blog | 📄 Paper 1. Model Introduction Kimi K2-Instruct-0905 is the latest, most capable version of Kimi K2. It is a state-of-the-art mixture-of-experts (MoE) language model, featuring 32 billion activated parameters and a total of 1Kimi09/05 00:00
Kimi K2: Open Agentic IntelligenceUpdate(0905) :We have an updated weight that enhances agentic coding and supports 256K context , see here Kimi K2 is our latest Mixture-of-Experts model with 32 billion activated parameters and 1 trillion total parameters. It achieves stateKimi07/11 00:00
Kimi-ResearcherMeet Kimi-Researcher , an autonomous agent that excels at multi-turn search and reasoning. It performs an average of 23 reasoning steps and explores over 200 URLs per task. Built on an internal version of the Kimi k-series model and trainedKimi06/20 00:00
Kimi-DevIntroducing Kimi-Dev: A Strong and Open-source Coding LLM for Issue Resolution Kimi-Dev Team 📄 Tech Report (Arxiv) | 🤗 Huggingface | 💻 Demo (HF Space) We introduce Kimi-Dev-72B, our new open-source coding LLM for software engineering tasKimi06/17 00:00
Kimi-AudioKimi-Audio-7B 🤗 | Kimi-Audio-7B-Instruct 🤗 | 📑 Paper We present Kimi-Audio, an open-source audio foundation model excelling in audio understanding, generation, and conversation . This repository contains the official implementation, modeKimi04/26 00:00
Kimina-Prover PreviewNumina & Kimi Team Tech Report | HuggingFace | 🧮 Demo | Kimina Lean Server | 🔢 CombiBench 🚀 UPDATE - Jul 10, 2025 We are excited to announce the official release of Kimina-Prover-72B ! For detailed information about this release, please Kimi04/15 00:00
Kimi-VLKIMI-VL TECHNICAL REPORT Tech Report | HuggingFace | 💬 Chat with Latest Kimi-VL (2506) 1. Introduction We present Kimi-VL , an efficient open-source Mixture-of-Experts (MoE) vision-language model (VLM) that offers advanced multimodal reasoKimi04/10 00:00
Muon is Scalable for LLM TrainingTech Report | HuggingFace | Megatron-LM Abstract Recently, the Muon optimizer based on matrix orthogonalization has demonstrated strong results in training small-scale language models, but the scalability to larger models has not been proveKimi02/23 00:00
MoBA: Mixture of Block Attention for Long-Context LLMsFull Report 🚀 Introducing MoBA --- Mixture of Block Attention Trainable Block Sparse Attention : The full context is divided into blocks, where each query token learns to attend to the most relevant KV blocks, enabling efficient processingKimi02/18 00:00
Kimi K1.5: Scaling Reinforcement Learning with LLMsKimi Team Full Report 🚀 Introducing Kimi k1.5 --- an o1-level multi-modal model Sota short-CoT performance, outperforming GPT-4o and Claude Sonnet 3.5 on 📐AIME, 📐MATH-500, 💻 LiveCodeBench by a large margin (up to +550%) Long-CoT performKimi01/20 00:00
Mooncake: A KVCache-centric Disaggregated Architecture for LLM ServingA KVCache-centric Disaggregated Architecture for LLM Serving Paper | Slides | Traces | Documentation | Blog | Slack Mooncake is the serving platform for Kimi , a leading LLM service provided by Moonshot AI . Under real workloads, Mooncake’sKimi06/26 00:00