
DeepSeek's V4.1 Flash Adds Live Web Search to Its 552B Agent Model
DeepSeek ships a snapshot refresh of its sparse MoE Flash model with cheaper caching, and opens its Search and Fetch APIs to all developers.
A focused feed of models, agents, research and open-source releases for people building with AI.

DeepSeek ships a snapshot refresh of its sparse MoE Flash model with cheaper caching, and opens its Search and Fetch APIs to all developers.

A community mashup stacks 40 persona modes on top of uncensored Qwen3.8-27B weights, runs on a single 24GB GPU, and ships ready for llama.cpp and LM Studio.

Meta Superintelligence Labs releases a 30B dense multimodal model, distilled from Muse Spark, that runs agentic workloads locally on 32GB consumer GPUs.

Kyutai's REKI is a 2.4B vision-language-action model trained only on public data, matching models twice its size with a cached memory trick.

Reflection AI opens the weights of Beam, a 501B Mixture-of-Experts model, alongside a 40+ partner ecosystem and a full agent toolchain.

Comfy-Org repackaged the OpenMOSS 5B diffusion transformer for ComfyUI, bringing JSON layout prompts and strong text rendering to local workflows under Apache 2.0.

Tencent Hunyuan open-sources a 30B memory writer model, a dual-system memory SDK, and MemHub to unify memory across AI tools.

Meta's open-source audio generation library bundles MusicGen, AudioGen, EnCodec, MAGNeT, JASCO and AudioSeal into one PyTorch toolkit for text-to-music research.

An open-source desktop canvas lets creators chain prompts, images, and video models into reusable AI storyboarding workflows, with assets stored locally.

Meta's open-source speech system covering 1,600+ languages, including 500 never before transcribed by AI, lands on TIME's Best Inventions list.

Meta's new 816B parameter mixture-of-experts model targets long-horizon coding and agentic work, with open weights that fit on two Mac Studios.

Perplexity released SPACE under Apache 2.0, the Firecracker-based microVM sandbox that isolates every agent task in Computer with external credential and egress control.

NVIDIA's new open 120B mixture-of-experts model fuses text, image, document, and video reasoning into one agent-ready backbone with 12B active parameters.

Mistral's first time-series foundation model family hits #1 on three major forecasting benchmarks with just 209M parameters and runs on CPU.

GVHMR recovers world-grounded 3D human motion from a single video using gravity-aligned coordinates, processing 45 seconds of footage in 280 ms.

IFM's new Arabic-first model family spans 0.9B to 375B parameters, with a flagship MoE that matches Claude Sonnet 5 on Arabic benchmarks and acts as an agent.

Black Forest Labs refreshes its FLUX 3 Image Klein workflows, pushing prompt upsampling and 2K rendering as the recommended path to top quality.

Zed v1.24 ships local Git tag creation, drag-and-drop terminal tabs into the Agent Panel, and smarter File Finder prefill from selections.

A community GGUF port of OpenMOSS's 8B music understanding model lets a C++/GGML desktop app caption your own audio dataset locally.

Google Research open-sourced Harness², a framework that auto-tunes the prompts, tools, and memory around frozen models to boost agent performance on unseen tasks.

Japan's National Institute of Informatics releases an 8B dense and 32B MoE model trained from scratch on 12T tokens, with the full pipeline open to inspection.

An open-source Mac menu bar app uses Google's EmbeddingGemma 2 to deliver fully local semantic search across photos, PDFs, audio, video and code.

Fish Audio's new 35B mixture-of-experts text-to-speech model ships open weights, with plain-language direction, mid-sentence voice shifts, and word-level audio inpainting.

Black Forest Labs released a 5.6B open-weights image model that handles text-to-image, editing, and multi-reference composition, running on 8GB GPUs.
No stories match these filters.