Thursday, July 23, 2026
- Tencent Yuanbaov4Tencent Yuanbao (Chinese: 腾讯元宝; pinyin: Téngxùn Yuánbǎo) is a consumer-facing artificial intelligence assistant application developed and operated by Chinese technology company Tencent.
- Doubao Seedancev3Doubao-Seedance is a family of video generation foundation models developed by ByteDance Seed, the AI research division of Chinese technology conglomerate ByteDance.
- Reve Imagev5Reve Image is a family of text-to-image generative models developed by Reve AI, Inc., a Palo Alto, California startup, whose current flagship, Reve 2.0 (released 3 June 2026)
- DALL-E 3v6DALL-E 3 (stylized by OpenAI as DALL·E 3) is the third generation of OpenAI's text-to-image system, announced on September 20, 2023 and released to ChatGPT Plus and Enterprise subscribers in October 2023
- HELM (Holistic Evaluation of Language Models)v4HELM (Holistic Evaluation of Language Models) is an open-source benchmark framework created by the Center for Research on Foundation Models (CRFM) at Stanford University for the reproducible and transparent…
- MMStarv4MMStar (Multi-modal Star) is a vision-language model evaluation benchmark consisting of 1,500 multimodal samples that were filtered from six pre-existing benchmarks to ensure both visual dependency (questions…
- Würstchenv4Würstchen is an efficient three-stage cascaded latent diffusion architecture for text-to-image synthesis introduced by Pablo Pernias, Dominic Rampas, Mats L. Richter, Christopher J. Pal
- Context Parallelismv4Context Parallelism (CP) is a distributed training strategy that partitions the input sequence dimension of a transformer across multiple accelerators and uses ring-style point-to-point communication to…
- Expert Parallelismv4Expert Parallelism (EP) is a model-parallelism strategy specific to Mixture of Experts (MoE) neural networks in which the individual expert sub-networks (typically feed-forward blocks) are sharded across…
- Sequence Parallelismv5Sequence parallelism (SP) is a family of distributed training techniques for transformer-based neural networks that partitions activations along the sequence (token) dimension across multiple accelerators…
- Qwen2-VLv4Qwen2-VL is a family of open-weight vision-language models released by the Qwen team at Alibaba Cloud between August and September 2024, in 2B, 7B, and 72B Instruct sizes.
- Llama 3.2 Visionv4Llama 3.2 Vision is the set of multimodal (image-plus-text) models in Meta's Llama 3.2 family, released on September 25, 2024 at the Meta Connect 2024 developer conference.
- InternVideov4InternVideo is a family of general-purpose video foundation models developed by OpenGVLab at the Shanghai Artificial Intelligence Laboratory in collaboration with Nanjing University and the Shenzhen Institutes…
- RFdiffusionv4RFdiffusion (short for RoseTTAFold diffusion) is a deep-learning system for de novo protein design developed at the University of Washington Institute for Protein Design (IPD) in David Baker's lab.
- Chai-1v4Chai-1 is an open multimodal foundation model for biomolecular structure prediction released by the San Francisco startup Chai Discovery on 9 September 2024.
- AlphaFold-Multimerv4AlphaFold-Multimer is a deep learning system for predicting the three-dimensional structures of protein complexes, released by Google DeepMind in October 2021 as an extension of AlphaFold 2.
- DeepSeek-Coderv4DeepSeek-Coder is a family of open-weight code large language models built for code generation, completion, and infilling, developed by the Chinese AI lab DeepSeek (DeepSeek-AI).
- DeepSeekMathv4DeepSeekMath is a family of open-weight large language models specialized for mathematical reasoning, released by Chinese AI laboratory DeepSeek in February 2024.
- DeepSeek-VL2v4DeepSeek-VL2 is an open-weights family of Mixture-of-Experts (MoE) vision-language models released by the Chinese AI laboratory DeepSeek on December 13, 2024.
- DeepSeek-Proverv5DeepSeek-Prover is a family of open-weight large language models developed by Chinese AI laboratory DeepSeek for formal theorem proving in the Lean 4 proof assistant.
- Graph of Thoughtsv3Graph of Thoughts (GoT) is a prompting and reasoning framework that models the intermediate steps of a large language model as an arbitrary directed graph rather than a linear chain or a tree.
- DeepSeek Janusv5DeepSeek Janus is a family of open-weight unified multimodal models from Chinese AI lab DeepSeek that perform both image understanding and text-to-image generation in a single autoregressive Transformer.
- Least-to-Most Promptingv3Least-to-Most Prompting is a few-shot prompting technique for large language models introduced by researchers at Google Brain in May 2022.
- Skeleton-of-Thoughtv4Skeleton-of-Thought (SoT) is a prompting technique for large language models that reduces end-to-end generation latency by first eliciting a short outline of the answer (the "skeleton") and then expanding each…
- Auto-CoTv4Auto-CoT (Automatic Chain of Thought) is an automated prompting method that builds few-shot Chain-of-Thought demonstrations for large language models without any human-written exemplars, introduced by…
- Chain of Density promptingv4Chain of Density (CoD) is a prompting technique for abstractive text summarization with large language models, introduced in the 2023 paper "From Sparse to Dense: GPT-4 Summarization with Chain of Density…
- Self-Refinev4Self-Refine is an inference-time prompting framework in which a single large language model iteratively improves its own output by alternating between generating natural-language feedback on a draft and…
- Step-Back Promptingv5Step-Back Prompting is a two-stage prompting technique introduced by researchers at Google DeepMind in October 2023.
- Pallas (JAX kernel language)v4Pallas is an experimental extension to JAX that lets users write custom hardware kernels in Python and lower them to both Tensor Processing Units and NVIDIA GPUs from a single source.
- Triton (OpenAI GPU programming language)v4Triton is an open-source, Python-embedded programming language and compiler, originally created by Philippe Tillet and released by OpenAI in July 2021
- Self-Discover promptingv4Self-Discover is a prompting framework for large language models in which the model first composes a task-specific reasoning structure from a library of atomic reasoning modules
- ThunderKittensv4ThunderKittens (often abbreviated TK) is an embedded C++ domain-specific language and header-only library for writing high-performance AI kernels on modern NVIDIA GPUs.
- torch.compilev3torch.compile is the just-in-time graph capture and compilation feature introduced in PyTorch 2.0, a release first announced at the PyTorch Conference on December 2, 2022 and shipped as a stable version on…
- tinygradv4tinygrad is an open-source deep learning framework written primarily in Python that aims to occupy the space between Andrej Karpathy's pedagogical micrograd and full-scale production stacks like PyTorch.
- Vast.aiv4Vast.ai is a cloud-based GPU rental marketplace that connects independent hardware operators (called hosts) with developers, researchers and companies looking to rent compute by the second.
- Ray Servev4Ray Serve is a scalable, framework-agnostic model serving library built on top of the Ray (framework) distributed computing system.
- Voltage Parkv4Voltage Park is a United States cloud computing company that operates a fleet of NVIDIA H100 graphics processing units for artificial intelligence training and inference workloads.
- OpenVINOv4OpenVINO (Open Visual Inference and Neural Network Optimization) is an open-source toolkit developed by Intel for optimizing and deploying deep learning inference across a heterogeneous set of Intel hardware…
- Text Generation Inference (TGI)v4Text Generation Inference (TGI) is an open-source toolkit developed by Hugging Face for deploying and serving large language models in production.
- Mike Lewisv4Mike Lewis is a British natural language processing researcher based in Seattle who serves as a research scientist at Meta AI (Facebook AI Research, FAIR) and as the pre-training research lead on the Llama…
- LiteLLMv4LiteLLM is an open-source AI gateway from BerriAI that lets developers call more than 100 large language model providers (including OpenAI, Anthropic, Google Gemini, Amazon Bedrock and Azure OpenAI) through a…
- BentoMLv4BentoML is an open-source Python framework for packaging, serving, and deploying machine learning and AI models as production inference services.
- Alec Radfordv4Alec Radford is an American machine learning researcher who was the first author of the papers introducing the DCGAN generative adversarial network, the original Generative Pre-Trained Transformer (GPT-1)…
- Greg Brockmanv4Greg Brockman is an American software engineer and entrepreneur who co-founded OpenAI in December 2015 and serves as its president and chairman of the board.
- Hybrid searchv4Hybrid search is an information retrieval technique that runs a lexical (sparse) keyword retriever, typically BM25, and a semantic (dense) vector retriever in parallel against the same corpus
- SEDD (Score Entropy Discrete Diffusion)v4Score Entropy Discrete Diffusion (SEDD) is a discrete diffusion model for language and other discrete data introduced by Aaron Lou, Chenlin Meng, and Stefano Ermon at Stanford University in the paper Discrete…
- LLaDA (Large Language Diffusion)v3LLaDA (Large Language Diffusion with mAsking) is a family of non-autoregressive large language models that generate text by iteratively denoising a sequence of mask tokens rather than predicting tokens left to…
- HyDE (Hypothetical Document Embeddings)v4HyDE (Hypothetical Document Embeddings) is a zero-shot dense retrieval technique that, instead of searching with the user's query, first prompts an instruction-following large language model to write a…
- SPLADEv5SPLADE (Sparse Lexical and Expansion model) is a learned sparse retrieval model that encodes a query or document as a weighted
- ColBERTv4ColBERT (Contextualized Late Interaction over BERT) is a neural information retrieval model that encodes queries and documents into matrices of token-level vectors and scores them with a late-interaction…
- Cross-encoderv4A cross-encoder is a neural network architecture that scores a pair of texts by feeding them jointly into a single transformer (such as BERT) and reading out one scalar score for the pair, for example a…
- Optimum-Quantov3Optimum Quanto, commonly referred to as Quanto, is a PyTorch-based quantization toolkit developed and maintained by Hugging Face that provides linear weight and activation quantization across multiple devices…
- NormalFloat 4-bit (NF4)v4NormalFloat 4-bit (NF4) is a 4-bit numerical data type for storing the weights of deep neural networks, introduced in the 2023 QLoRA paper by Tim Dettmers, Artidoro Pagnoni, Ari Holtzman, and Luke Zettlemoyer…
- FP4 (4-bit floating point)v4FP4 (4-bit floating point) is a numerical format that stores a real number in just 4 bits, the smallest floating-point type in mainstream use for deep learning.
- SmoothQuantv5SmoothQuant is a training-free, accuracy-preserving post-training quantization (PTQ) method that enables 8-bit weight and 8-bit activation (W8A8) integer inference for large language models without retraining…
- Lookahead Decodingv4Lookahead Decoding is a parallel decoding algorithm for accelerating inference in large language models, introduced in November 2023 by Yichao Fu, Peter Bailis, Ion Stoica, and Hao Zhang from the Hao AI Lab at…
- ExLlamaV2 (EXL2)v4EXL2 (ExLlamaV2 format) is an open-source, mixed-bit weight-quantization format for compressing large language models so they run fast on a single consumer-class NVIDIA GPU.
- YOCO (You Only Cache Once)v4YOCO ("You Only Cache Once") is a decoder-decoder neural network architecture for large language models introduced by researchers at Microsoft Research and Tsinghua University in May 2024.
- EAGLE-2v4EAGLE-2 ("Faster Inference of Language Models with Dynamic Draft Trees") is the second generation of the EAGLE family of speculative decoding methods for accelerating large language model inference, introduced…
- Rule-Based Rewards (RBR)v4Rule-Based Rewards (RBR) is a safety-alignment technique introduced by OpenAI in July 2024 that replaces large quantities of human-labeled safety preference data with an explicit collection of natural-language…
- LIMA (Less Is More for Alignment)v3LIMA, short for "Less Is More for Alignment," is a 2023 research paper by Chunting Zhou and colleagues at Meta AI, Carnegie Mellon University, the University of Southern California, and Tel Aviv University…
- Self-Rewarding Language Modelsv6Self-Rewarding Language Models (SRLM) is an iterative alignment method in which a single large language model alternately plays the role of policy (generating candidate responses to user prompts) and reward…
- BrainChip Akidav5Akida is a family of fully digital, event-based neuromorphic processor intellectual property (IP) cores and reference chips developed by BrainChip Holdings Ltd
- Intel Loihiv4Intel Loihi is a family of research neuromorphic processors developed by Intel Labs to implement spiking neural networks (SNNs) in silicon, with the stated goals of energy-efficient inference, online learning…
- RoseTTAFoldv4RoseTTAFold is a deep-learning system for predicting the three-dimensional structures of proteins from amino acid sequence, developed at the Institute for Protein Design (IPD) at the University of Washington…
- Untether AIv4Untether AI Corp. was a Toronto-based fabless semiconductor startup that designed energy-efficient inference accelerators based on a proprietary "at-memory compute" architecture, in which large arrays of small…
- DatologyAIv4DatologyAI is a Redwood City, California artificial-intelligence startup that builds automated tools for curating, deduplicating, and composing the training datasets used by foundation models.
- Surge AIv5Surge AI (legal entity Surge Labs Inc., often stylized SurgeHQ) is an American data annotation and human evaluation company headquartered in San Francisco, California that supplies frontier artificial…
- Backdoor attacks on large language modelsv4A backdoor attack on a large language model (LLM) is an adversarial training-time attack in which an attacker manipulates training data, fine-tuning data, preference labels, or model weights so that the…
- Essential AIv5Essential AI is a San Francisco based artificial intelligence research company founded in 2023 by Ashish Vaswani and Niki Parmar
- Imbuev5Imbue is a San Francisco artificial intelligence research lab focused on training foundation models and building agent systems oriented toward reasoning and software engineering.
- Membership Inference Attackv4A Membership Inference Attack (MIA) is a privacy attack against a trained machine learning model in which an adversary, given a candidate data record and access to the model
- Trusted Execution Environments for machine learningv4Trusted Execution Environments for machine learning (TEEs for ML, sometimes marketed as "Confidential AI" or "confidential inference") are deployments of hardware-isolated execution environments to run…
- Jan Leikev4Jan Leike is a German machine learning researcher who specializes in artificial intelligence alignment and, since May 2024, leads the Alignment Science team at Anthropic.
- Evan Hubingerv4Evan Hubinger is an American AI safety researcher who leads the alignment stress-testing team at Anthropic, where he serves as a Member of Technical Staff and manager.
- Shane Leggv4Shane Legg (born 1973) is a New Zealand-born machine learning researcher and entrepreneur who co-founded the artificial intelligence laboratory DeepMind in 2010 with Demis Hassabis and Mustafa Suleyman
- Homomorphic encryption for machine learningv4Homomorphic encryption for machine learning is the application of fully, somewhat, or leveled homomorphic encryption (FHE, SHE, LHE) so that a server can run machine learning computations, almost always…
- Indirect prompt injectionv4Indirect prompt injection is a class of attack against large language model-integrated applications in which the malicious instructions that subvert the model are not supplied by the user, but are smuggled…
- WizardLMv3WizardLM is a family of open-weights instruction-tuned LLaMA-derived large language models and an associated data-synthesis methodology, both produced by a research group at Microsoft led by Can Xu.
- Self-Instructv5Self-Instruct is a semi-automated framework for aligning a pretrained large language model with instruction-following behavior by bootstrapping its training data from the model itself, introduced in December…
- Model extraction attackv5A model extraction attack is a class of machine learning security attacks in which an adversary, restricted to black-box query access to a target model (typically through a paid prediction API)
- OpenOrcav4OpenOrca is a large open-source instruction-tuning dataset that augments the FLAN Collection with chain-of-thought responses generated by OpenAI's GPT-3.5 and GPT-4 APIs.
- UltraChatv4UltraChat is a large-scale synthetic multi-turn instructional conversation dataset released in May 2023 by the OpenBMB group at Tsinghua University, comprising approximately 1.5 million dialogues generated by…
- The Stack (BigCode dataset)v4The Stack is a family of large, permissively-licensed source-code datasets built by the BigCode project, an open scientific collaboration jointly led by Hugging Face and ServiceNow Research, to train and…
- SlimPajamav4SlimPajama is a 627-billion-token English-language pre-training corpus for large language models, produced by Cerebras Systems in collaboration with the Opentensor Foundation by extensively cleaning and…
- Cybenchv5Cybench (short for Cybersecurity benchmark) is an open-source evaluation framework for measuring the cybersecurity capabilities and risks of large language model agents.
- LTX-Videov4LTX-Video is an open-source, transformer-based latent video diffusion model developed by the Israeli company Lightricks and first released to the public in November 2024.
- Lumierev4Lumiere is a text-to-video diffusion model developed by Google Research in collaboration with researchers from the Weizmann Institute of Science, Tel Aviv University, and the Technion
- SIMA (DeepMind)v4SIMA (Scalable Instructable Multiworld Agent) is a family of generalist embodied AI agents developed by Google DeepMind that follow free-form natural-language instructions to act in a wide range of…
- Open-Sorav4Open-Sora is an open-source text-to-video diffusion project initiated in March 2024 by Singapore-based startup HPC-AI Tech (the team behind the Colossal-AI distributed training framework) as a public attempt…
- Goodhart's lawv6Goodhart's law states that "when a measure becomes a target, it ceases to be a good measure": any statistical regularity or metric tends to break down once it is used as a target for control or decision-making.
- Stable Video Diffusionv4Stable Video Diffusion (SVD) is a latent video diffusion model released by Stability AI on 21 November 2023
- Open Interpreterv4Open Interpreter is an open-source desktop agent, distributed as a Python command-line tool and library, that lets a large language model write and execute code locally on the user's machine.
- Guidance (library)v4Guidance is an open-source Python library, originally developed at Microsoft Research, for building structured, multi-step programs that drive large language models.
- Outlines (library)v4Outlines is an open-source Python (programming language) library, released under the Apache 2.0 license, that constrains large language model output to user-specified structures: regular expressions, function…
- Mochi 1v4Mochi 1 is an open-weights text-to-video diffusion model released by Genmo Inc. on October 22, 2024 under the Apache 2.0 license.
- StreamingLLMv4StreamingLLM is an inference-time technique that allows pretrained transformer language models, originally trained with a finite attention window
- H2O (Heavy-Hitter Oracle for KV Cache)v4H2O (Heavy-Hitter Oracle) is a training-free, runtime KV cache eviction policy for autoregressive large language model inference.
- Instructor (library)v4Instructor is an open-source Python library that returns type-safe, Pydantic-validated structured outputs from large language model APIs.
- ChatDevv3ChatDev is an open-source multi-agent software development framework in which large language model (LLM) agents play role-specialized parts (Chief Executive Officer, Chief Technology Officer, programmer…
