Trending AI Models
Live tracker for trending ai models. Updated daily from five source platforms.
About this hub
New AI models drop almost weekly — base models, fine-tunes, multi-modal variants, reasoning extensions, distilled small versions. Most coverage focuses on the household names (GPT, Claude, Gemini) and misses the second tier where the actual innovation is happening: the open-weight finetunes outperforming closed APIs on niche tasks, the regional labs shipping competitive models, the architectures the next research wave will build on. This page tracks the model launches and re-launches that are actually gaining mindshare — measured by cross-source mention growth on HN, Reddit ML communities, GitHub Trending, and ML Twitter. We don't list every model; we list the ones being talked about right now.
Top trending this week
24 trending ai models ranked by signal velocity
Claude Opus 4.7
Anthropic's latest flagship model with top-tier reasoning and agentic coding performance
Open reportThinking Machines Lab
A multimodal AI lab redefining interaction beyond turn-based conversation.
Open reportVeo 3
Google's latest video generation model producing cinema-quality footage from text prompts.
Open reportRay2
Luma Labs' new video model generating coherent minute-long clips from text prompts.
Open reportKrea V3
Real-time image and video generation model with instant preview and editing
Open reportNeedle
A compact 26M parameter model for efficient tool calling, distilled from Gemini
Open reportKling 2
A next-gen video generation model with physics simulations that mimic real life.
Open reportKimi K2.5
A 1-trillion-parameter open-weights model from Moonshot AI for advanced reasoning tasks
Open reportOpenAI Voice Trio
Three specialized OpenAI models for real-time voice reasoning, translation, and transcription
Open reportMythos
An AI model for autonomous vulnerability discovery that challenges IMF classification norms.
Open reportNvidia World Model
A 2.6B parameter model that generates interactive 3D worlds from a single image
Open reportDeepSeek R1
Open-source reasoning model from China matching OpenAI o1 performance
Open reportNousCoder-14B
Open-source 14B parameter coding model from Nous Research, competitive with larger models
Open reportHeaviside
A foundation model for electromagnetism trained on proprietary physics data
Open reportMiniCPM5-1B
A compact open model achieving state-of-the-art performance for edge deployment
Open reportLuma Ray
Luma AI's video model update improving character consistency across frames
Open reportDeepSeek-V4-Flash
A new DeepSeek model that enables advanced LLM steering through vector manipulation.
Open reportGemini Omni
A multimodal AI model that generates content from any input, starting with video.
Open reportNeuraCode
A specialized coding model that outperforms leaders on agentic benchmarks.
Open reportRealVoice 2
OpenAI's real-time voice model for natural speech and translation
Open reportGPT-5.5
OpenAI's gated GPT-5.5 model with a specialized cybersecurity variant
Open reportOpen R1
Open source reasoning model that rivals closed models on key benchmarks
Open reportSuno v4
AI music generation model that creates full songs with coherent structure
Open reportR1 Distill
A distilled version of DeepSeek R1 that matches larger models with fewer parameters
Open reportNew entries today
8 additions in the last 7 days
Nvidia World Model
A 2.6B parameter model that generates interactive 3D worlds from a single image
Open reportDeepSeek R1
Open-source reasoning model from China matching OpenAI o1 performance
Open reportMiniCPM5-1B
A compact open model achieving state-of-the-art performance for edge deployment
Open reportGemini Omni
A multimodal AI model that generates content from any input, starting with video.
Open reportQwen3.7-Max
Alibaba's latest large language model optimized for agentic tasks and tool use
Open reportGemini 3.5 Flash
Google's fast frontier model with enhanced action capabilities for developers
Open reportAVTR-1 Real-Time Open Weights Model
Open-weight model for generating uncanny AI avatars in real time
Open reportGPT-5
OpenAI's next-generation flagship language model with advanced reasoning and multimodal capabilities.
Open reportTrack this vertical daily.
Subscribe for a morning digest of trending ai models movement.