Summaries

By ByteDance

Seedance 2.0 is ByteDance's multimodal AI video model that turns text plus image, video and audio references into high-resolution, sound-synced clips, giving creators director-level control over camera, motion, style and multi-shot storytelling.

🎥Videos 🖼️Image generation 📚Summaries

Multimodal

Released 4mo ago

Gen 4

XMAX X1

By Xmax AI

XMAX X1 is a real-time interactive video model that fuses virtual and real worlds, using the phone camera and touch gestures for millisecond-level, on-device AR-style experiences.

🎥Videos 📢Prompts 📚Summaries 📚Novels

Video

Released 4mo ago

Gen 7

K EXAONE

By LG Electronics

K-EXAONE is LG AI Research’s 236B-parameter MoE language model with 23B active experts, 256K context and strong multilingual reasoning, built as a national Korean foundation model with efficient hybrid attention.

📝Writing 📚Summaries 📚Learning 🚀Productivity

Text

Released 5mo ago

Gen 4

Chenkin Noob XL

By ChenkinNoob Team

Chenkin Noob XL (CKXL) is a Stable Diffusion XL-class anime image checkpoint built on the Noob XL / noobai-XL-1.1 backbone, retrained on an extended Danbooru dataset through late 2025 to improve texture, lighting and style consistency.

🎨Anime images 📚Summaries 📷Images 🎨Cartoon images

Image

Released 5mo ago

Gen 7

VibeThinker-1.5B

By WeiboAI

VibeThinker-1.5B is WeiboAI’s compact instruction model designed for fast, cost-efficient assistants and automation.

📝Writing 📚Summaries 💬Chatting

Text

Released 7mo ago

Gen 7 DeepSeek

DeepSeek V3.1 Terminus

By DeepSeek

DeepSeek-V3.1-Terminus is DeepSeek’s flagship reasoning model, tuned for difficult analysis, math, and coding. It supports very long context, function/tool calling, reliable JSON outputs, and an optional extended-thinking mode—ideal for enterprise RAG, agents, and high-stakes workflows.

📊Data analysis 💻Coding 📚Summaries

Text

Released 8mo ago

Gen 3 ERNIE

ERNIE 4.5 Turbo

By Baidu

ERNIE 4.5 Turbo is Baidu’s high-throughput, cost-optimized variant of ERNIE 4.5. It delivers strong reasoning and coding with long-context options, tool/function calling, JSON outputs, and streaming—ready for production via ERNIE Bot and the Qianfan API.

💻Coding 📚Summaries 🚀Productivity

Text

Released 8mo ago

Gen 7 Ling

Ling mini 2.0

By AntGroup

Ling-mini-2.0 is a compact multilingual instruction model tuned for speed, cost efficiency, and dependable reasoning. It supports long context, tool and function calling, and clean JSON outputs, making it a solid fit for everyday copilots, chatbots, and lightweight automation.

💻Coding 📚Summaries 🤖Chatbots 🎥Videos

Text

Released 9mo ago

Gen 4

Dia 1.6B

By Nari Labs

Dia-1.6B is Nari Labs’ ultra-compact instruction model—about 1.6B parameters—built for fast, low-cost deployment on a single GPU or edge device. It handles everyday reasoning, summarization, and lightweight coding, with structured JSON outputs and tool/function calling for agents.

📚Summaries 💻Coding 📝Writing 💬Chatting

Audio

Released 1y ago

Gen 7 LLaMA

Llama 4 Scout

By Meta Platforms

Large Multimodal Model; MoE architecture (16 experts, 17B active, 109B total params), industry-le...

✉️Marketing emails 📝Writing 📢Prompts 📚Summaries

Text

Released 1y ago

Gen 3 Qwen

Qwen2.5 Omni

By Alibaba

Qwen2.5-Omni is an end-to-end multimodal model designed to perceive diverse modalities, including text, images, audio, and video, while simultaneously generating text and natural speech responses in a streaming manner.

📚Learning 💻Coding 📚Summaries 🎥Videos

Multimodal

Released 1y ago

Gen 3 Mistral

Mistral Small 3.2

By Mistral AI

Mistral Small 3.2 (24B) is the largest configuration of the “Small” line—around 24B dense parameters—offering stronger reasoning and coding skills while staying cheaper and faster than Mistral Medium or Large. It supports long context, JSON outputs, and tool/function calling for production copilots and RAG.

💻Coding 📝Writing 📚Summaries 🎥Videos

Text

Released 1y ago

Gen 7 Command

Command A

By Caldera Labs

Cohere Command A is an enterprise-grade instruction model built for reliable tool use, RAG, and multilingual workflows. It delivers strong reasoning and coding with long-context support, structured (JSON) outputs, and straightforward deployment in production stacks.

💻Coding 📊Data analysis 📝Writing 📚Summaries

Text

Released 1y ago

Gen 3

Goku

By Saiyan World

Goku is a family of joint image-and-video generation models based on rectified flow Transformers, supporting text-to-video, image-to-video, and text-to-image generation.

🖼️Image generation 📚Summaries 📷Images 🎥Videos

Multimodal

Released 1y ago

Gen 3 Nemotron

Cosmos Nemotron VLM

By NVIDIA

Cosmos Nemotron VLM is NVIDIA’s multimodal model that fuses Cosmos world-model perception with Nemotron language reasoning. It understands images and video alongside text, performs step-by-step visual reasoning, and supports tool/function calling and JSON outputs—optimized for fast, scalable deployment via TensorRT-LLM and NIM.

📚Summaries 📝Writing 📷Images 🎥Videos

Text

Released 1y ago

Gen 7 Nova

Nova Micro

By Amazon

Nova Micro is the compact tier of the Nova family, tuned for low-latency assistants and high-throughput apps. It follows instructions cleanly, supports long context, tool and function calling, and reliable JSON output, with steady multilingual and lightweight coding performance.

💬Chatting 📚Summaries 🤖ChatGPT 🎥Presentations

Text

Released 1y ago

Gen 7 Qwen

Qwen 2.5 Turbo

By Alibaba

Qwen 2.5 Turbo is the low-latency tier for high-throughput apps. It offers solid reasoning and coding with long context, tool calling, and reliable JSON at lower cost.

💻Coding 📝Writing 📚Summaries 💬Chatting

Text

Released 1y ago

Gen 7

Aya Expanse

By Caldera Labs

Aya Expanse is a multilingual, reasoning-focused LLM designed for high-quality instruction following, translation, and coding across many languages. It supports long-context prompts, tool/function calling, and reliable JSON outputs—well suited for RAG, agents, and enterprise copilots.

🌐Text translation 💻Coding 📚Summaries

Text

Released 1y ago

Gen 7 Kanana

Kanana Essence

By Kakao

Kanana Essence is a compact variant in Kakao’s Kanana model line, optimized for efficiency and everyday reasoning tasks. It balances smaller size with reliable instruction following, making it suitable for lightweight copilots, chat assistants, and cost-sensitive deployments.

💬Chatting 📞Customer support 📚Summaries 🎥Videos

Text

Released 1y ago

Gen 7 Mistral

Ministral 3B

By Mistral AI

Ministral 3B (2410) is Mistral’s ultra-compact dense LLM—around 3 billion parameters—built for speed, efficiency, and low compute cost. It supports instruction following, summarization, reasoning, and lightweight coding tasks, with JSON outputs and function/tool calling for agents and automations.

📚Summaries 📝Writing 📚Courses

Text

Released 1y ago

Gen 7 Mistral

Ministral 8B

By Mistral AI

Ministral 8B is Mistral’s compact dense LLM—about 8 billion parameters—built for fast, efficient reasoning, summarization, and coding tasks. It supports long-context prompts, structured JSON outputs, and tool/function calling, making it practical for cost-sensitive copilots and automation.

💻Coding 📝Writing 📚Summaries 🏋️Fitness

Text

Released 1y ago

Gen 7 Cerebras

Cerebras 8B

By Cerebras

Cerebras 8B is a compact general model for fast, low-cost chat, reasoning, and light coding that fits comfortably on a single modern GPU.

💬Chatting 💻Coding 📚Summaries 🔎Search

Text

Released 1y ago

Gen 7

Jamba 1.5 Mini

By AI21 Labs

Jamba 1.5 Mini is AI21’s long-context instruction model built on a hybrid SSM-Transformer architecture, optimized for high-throughput reasoning and generation on very long inputs.

📚Summaries 🔍SEO content 🤯Zizek debates

Text

Released 1y ago

Gen 7 Phi

Phi 3.5 mini

By Microsoft

Phi-3.5-mini is Microsoft’s compact open-weight LLM (~3.8B parameters) tuned for strong instruction following, reasoning, and coding. It supports long-context (up to 128K), function calling, and JSON outputs, and is MIT-licensed for easy use on Azure AI or locally.

💻Coding 📚Summaries 🏋️Fitness 💬Chatting

Text

Released 1y ago

Gen 7 Phi

Phi 3.5 MoE

By Gentext Group

Phi-3.5-MoE is Microsoft’s open-weight Mixture-of-Experts model (16×3.8B, ~6.6B active) built for strong reasoning and coding with a 128K-token context window. It’s MIT-licensed and available in Azure AI Foundry and on Hugging Face

💻Coding 📝Writing 📚Summaries 💬Chatting

Text

Released 1y ago

Gen 7

Titan Text Premier

By Amazon

Amazon Titan Text Premier is AWS’s top-tier text generation model on Bedrock, built for enterprise-grade reasoning, coding, and long-context tasks. It supports tool/function calling, reliable JSON output, streaming, and tight integration with Bedrock’s retrieval and guardrails.

📝Writing 📊Data analysis 💻Coding 📚Summaries

Text

Released 2y ago

Gen 7 Phi

Phi-3-small

By Microsoft

Phi-3 Small is Microsoft’s ~7B open-weight LLM tuned for instruction following, reasoning, and coding. It’s compact enough for single-GPU/edge deployment and supports tool/function calling and structured (JSON) outputs under a permissive license.

💻Coding 📝Writing 📚Summaries 💬Chatting

Text

Released 2y ago

Gen 5 Mistral

Mistral 7B

By Mistral AI

Mistral 7B is an open 7-billion parameter model built for strong quality at small size. It runs fast on a single modern GPU, follows instructions well in the Instruct variant, and delivers solid coding, summarization, and chat performance for self-hosted assistants.

💻Coding 📝Writing 📚Summaries 🤖ChatGPT

Text

Released 2y ago