Large Language Models

Models 15

• Gen 3
  ZAYA1-8B

  By Zyphra AI

  ZAYA1-8B is Zyphra’s small mixture-of-experts reasoning model, built for long-form reasoning, mathematics, and coding. It has 8.4B total parameters with 760M active parameters, is post-trained for reasoning, and is positioned as highly intelligence-efficient enough for both strong test-time-compute use and local on-device deployment.

  📚Large Language Models🔎Problem solving💻Coding🔢Math
  NewMultimodal

  Released 29d ago
• Gen 3
  SubQ 1M Preview

  By Subquadratic

  SubQ 1M-Preview is Subquadratic’s first large language model and the first fully subquadratic LLM, built so compute grows linearly with context length rather than quadratically. It is designed for massive-context reasoning, coding, search, and full-corpus processing, making million-token windows more practical, faster, and cheaper than standard transformer-based approaches

  📚Large Language Models🧠AI inference🦾Ai research
  NewMultimodal

  Released 30d ago
• Gen 7
  nanowhale

  By Hugging Face

  nanowhale is Hugging Face’s miniature DeepSeek-V4-style language model project, built to reproduce the core DeepSeek-V4 architecture at about 110M parameters. It is designed as an educational and research-scale model, with both a pretrained base version and an SFT chat version, rather than as a production frontier assistant.

  📚Large Language Models🧠Model training🦾Ai research
  NewText

  Released 1mo ago
• Gen 3 DeepSeek
  DeepSeek V4 Flash

  By DeepSeek

  DeepSeek-V4-Flash is DeepSeek’s faster, smaller, and more economical V4 model for efficient large-scale use. It has 284B total parameters with 13B active parameters, supports a 1M token context window, and is positioned as a high-speed, cost-effective model whose reasoning comes close to V4-Pro.

  📚Large Language Models💬Chatting💻Conversational coding🧠AI inference
  NewMultimodal

  Released 1mo ago
• Gen 3 DeepSeek
  Deepseek V4 Pro

  By DeepSeek

  DeepSeek-V4-Pro is DeepSeek’s new flagship open-source model for top-end reasoning, coding, world knowledge, and agentic work. It uses a MoE architecture with 1.6T total parameters and 49B active parameters, supports a 1M token context window, and is positioned as rivaling leading closed-source models while leading current open models on several major capabilities.

  📚Large Language Models💬Chatting💻Conversational coding🧠AI inference
  NewMultimodal

  Released 1mo ago
• Gen 3
  Ling 2.6 1T

  By AntGroup

  Ling-2.6-1T is inclusionAI’s trillion-parameter flagship instruct model for real-world agents. It is positioned for fast execution, high efficiency at scale, advanced coding, complex reasoning, and large-scale agent workflows, using a “fast thinking” approach to cut cost while keeping top-tier performance.

  📚Large Language Models💬Chatting💻Conversational coding
  NewMultimodal

  Released 1mo ago
• Gen 7
  F pre 8kv 8k 14k SWA

  By Ai2

  F_pre_8kv_8k_14k_SWA is an AllenAI OlmPool architectural-variation model for studying long-context extension. It is a 7B to 8B class language model trained to 150B tokens total, including 140B pretraining and 10B context extension, and is meant for research on architecture choices rather than polished instruction-following use.

  📚Large Language Models🔄Language model optimization
  NewText

  Released 1mo ago
• Gen 7
  Parcae

  By Sandy Research

  Parcae is a Sandy Research model family described on Hugging Face as “a family of stable looped models.” The public collection currently lists four sizes: 140M, 370M, 770M, and 1.3B.

  📚Large Language Models🔨LLM development🎓LLM training
  NewText

  Released 1mo ago
• Gen 7
  LFM2.5 350M

  By Liquid AI

  LFM2.5-350M is Liquid AI’s compact 350M-parameter language model optimized for fast inference, tool use, data extraction, and structured outputs. It is designed to run across cloud GPUs, CPUs, and edge devices, and is positioned for lightweight agentic workflows and large-scale data processing rather than math, coding, or creative writing.

  📚Large Language Models🔍Data extraction🔄Language model optimization🧠AI inference
  NewText

  Released 2mo ago
• Gen 3
  Xiaomi MiMo V2 Omni

  By Xiaomi

  MiMo-V2-Omni is an omni foundation model that unifies multimodal understanding with agentic capability, built to see, hear, and act.

  📚Large Language Models🎥Video summaries📚Audio summaries🔍Image interpretation
  NewMultimodal

  Released 2mo ago
• Gen 3
  Xiaomi MiMo V2 Pro

  By Xiaomi

  MiMo-V2-Pro is Xiaomi’s flagship foundation model built for real-world agent workloads, designed to act as the “brain” of agent systems that orchestrate complex workflows and tool use.

  📚Large Language Models🤖Agents🔄Workflow optimization💻Conversational coding
  NewText

  Released 2mo ago
• Gen 7
  Nemotron 3 Super

  By NVIDIA

  Nemotron 3 Super is NVIDIA’s open Mixture-of-Experts hybrid Mamba-Transformer model designed for high-throughput agentic workloads and long-context reasoning.

  🤖Agents📚Large Language Models🔍Advanced reasoning💻Conversational coding
  NewText

  Released 2mo ago
• Gen 7
  Sequential Hidden Decoding 8B n8

  By Tencent

  Sequential-Hidden-Decoding-8B-n8 is Tencent’s base language model variant that extends effective sequence length by 8x using multiple embedding passes while keeping the same Transformer backbone. Built on Qwen3-8B-Base, it is aimed at benchmarking, text completion, and fine-tuning research, with hidden latent steps acting like implicit reasoning before each predicted token.

  📚Large Language Models🤖Ai research assistance🔄Language model optimization🧠AI inference
  NewText

  Released 2mo ago
• Gen 3
  Covenant 72B

  By Covenant AI

  Covenant-72B is a permissionless collaboratively trained 72B-parameter language model trained from scratch via decentralized coordination on Bittensor infrastructure.

  📚Large Language Models💬Chatting🔍Advanced reasoning
  NewText

  Released 2mo ago
• Gen 3 GPT
  GPT 5.3

  By OpenAI

  GPT-5.3 is an updated general-purpose chat model optimized for smoother everyday conversations, more accurate answers, and better contextual understanding. It is positioned as a broadly useful assistant model that reduces unnecessary refusals, excessive caveats, and overly cautious phrasing, while keeping strong performance for common conversational and help-oriented tasks.

  📚Large Language Models💬Chatting🎙️Conversation enhancement🔍Conversational queries
  Multimodal

  Released 3mo ago