https://theopenweights.com The daily record of open-source AI — new model releases and major updates. en-us https://theopenweights.com/news/glm-5-2-bs32 https://theopenweights.com/news/glm-5-2-bs32 Wed, 17 Jun 2026 03:19:15 GMT Zhipu AI has released GLM-5.2, a new bilingual Mixture of Experts language model with sparse attention, available under the permissive MIT license. https://theopenweights.com/news/vibethinker-3b-3am6 https://theopenweights.com/news/vibethinker-3b-3am6 Fri, 12 Jun 2026 11:43:29 GMT Weibo AI has released VibeThinker-3B, a compact 3B model specializing in complex reasoning tasks like math, code generation, and GPQA benchmarks. https://theopenweights.com/news/kimi-k2-7-code-jmbg https://theopenweights.com/news/kimi-k2-7-code-jmbg Thu, 11 Jun 2026 07:51:47 GMT Moonshot AI has released Kimi-K2.7-Code, a Mixture-of-Experts model designed for coding tasks that can also process visual inputs like images. https://theopenweights.com/news/zonos-2-u7e2 https://theopenweights.com/news/zonos-2-u7e2 Thu, 11 Jun 2026 02:20:46 GMT Zyphra has released Zonos 2, an open-weight text-to-speech model available for commercial use under the permissive Apache 2.0 license. https://theopenweights.com/news/diffusiongemma-26b-a4b-h4ke https://theopenweights.com/news/diffusiongemma-26b-a4b-h4ke Tue, 09 Jun 2026 12:40:12 GMT Google has released DiffusionGemma 26B, an open-source, instruction-tuned Mixture-of-Experts model that uses a novel diffusion architecture for text. https://theopenweights.com/news/scail-2-xvt5 https://theopenweights.com/news/scail-2-xvt5 Tue, 09 Jun 2026 07:18:27 GMT Zhipu AI's research arm has released SCAIL-2, an open-source, MIT-licensed diffusion model for pose-driven character animation from a single image. https://theopenweights.com/news/north-mini-code-1-0-wul0 https://theopenweights.com/news/north-mini-code-1-0-wul0 Fri, 05 Jun 2026 02:38:01 GMT Cohere has released North-Mini-Code 1.0, an Apache 2.0-licensed Mixture-of-Experts model specialized for code generation and agentic chat applications. https://theopenweights.com/news/higgs-audio-v3-tts-4b-2be3 https://theopenweights.com/news/higgs-audio-v3-tts-4b-2be3 Thu, 04 Jun 2026 05:37:32 GMT Boson AI has released Higgs Audio v3, a 4-billion-parameter text-to-speech model for expressive, controllable, and multilingual audio generation. https://theopenweights.com/news/minimax-m3-1lji https://theopenweights.com/news/minimax-m3-1lji Tue, 02 Jun 2026 07:49:31 GMT MiniMax AI has released MiniMax-M3, an open-weight, multimodal Mixture-of-Experts model designed for vision, coding, and agentic reasoning tasks. https://theopenweights.com/news/ideogram-4-0-abj7 https://theopenweights.com/news/ideogram-4-0-abj7 Sat, 30 May 2026 23:40:55 GMT Ideogram has released Ideogram 4.0, a 9.3B parameter open-weight text-to-image model using a Diffusion Transformer (DiT) architecture. https://theopenweights.com/news/gemma-4-12b-lou5 https://theopenweights.com/news/gemma-4-12b-lou5 Sat, 23 May 2026 01:17:15 GMT Google DeepMind has released Gemma 4 12B Instruct, a new open-weight model with a unified 'any-to-any' architecture for flexible multimodal reasoning. https://theopenweights.com/news/sana-wm-bidirectional-3rui https://theopenweights.com/news/sana-wm-bidirectional-3rui Mon, 18 May 2026 11:14:09 GMT NVIDIA has released SANA-WM, an open-source world model for camera-controllable video generation using a novel bidirectional diffusion technique. https://theopenweights.com/news/nemotron-3-5-asr-streaming-0-6b-jn8s https://theopenweights.com/news/nemotron-3-5-asr-streaming-0-6b-jn8s Fri, 15 May 2026 21:52:25 GMT NVIDIA has released Nemotron 3.5 ASR Streaming, a 600M-parameter model designed for real-time, multilingual speech-to-text transcription. https://theopenweights.com/news/lance-8jr6 https://theopenweights.com/news/lance-8jr6 Fri, 15 May 2026 08:05:57 GMT ByteDance has released Lance, a 3B parameter open-source model for unified image and video generation, editing, and understanding via a single architecture. https://theopenweights.com/news/sensenova-u1-8b-mot-infographic-e0f6 https://theopenweights.com/news/sensenova-u1-8b-mot-infographic-e0f6 Thu, 14 May 2026 09:54:29 GMT SenseTime has released SenseNova U1, an 8B 'any-to-any' multimodal model. It specializes in generating and editing complex infographics within a conversation. https://theopenweights.com/news/ltx-2-3-22b-ic-lora-lipdub-azlx https://theopenweights.com/news/ltx-2-3-22b-ic-lora-lipdub-azlx Mon, 11 May 2026 10:02:08 GMT Lightricks has released an IC-LoRA, a specialized adapter for its LTX-2.3 video model, designed to enable high-quality, identity-preserving lip-dubbing. https://theopenweights.com/news/hunyuan-mt2-1-8b-pb3x https://theopenweights.com/news/hunyuan-mt2-1-8b-pb3x Mon, 11 May 2026 08:22:10 GMT Tencent has released Hunyuan-MT2, a 1.8B parameter dense model focused on high-quality multilingual machine translation, available with a custom license. https://theopenweights.com/news/supertonic-3-47h3 https://theopenweights.com/news/supertonic-3-47h3 Wed, 06 May 2026 20:46:20 GMT Supertone has released Supertonic 3, a multilingual text-to-speech model optimized for on-device use with ONNX support and a non-commercial license. https://theopenweights.com/news/nvidia-pid-pixel-diffusion-decoder-38mh https://theopenweights.com/news/nvidia-pid-pixel-diffusion-decoder-38mh Tue, 28 Apr 2026 00:41:46 GMT NVIDIA has released PiD, a pixel-diffusion VAE decoder for super-resolution. The component is designed to work with Stability AI's Z-Image base model. https://theopenweights.com/news/nemotron-3-nano-omni-30b-a3b-reasoning-cbno https://theopenweights.com/news/nemotron-3-nano-omni-30b-a3b-reasoning-cbno Fri, 24 Apr 2026 16:09:58 GMT NVIDIA has released Nemotron-3 Nano Omni, a 30B parameter Mixture-of-Experts model for multimodal reasoning that uses only 3B active parameters.