llms.blog
Models12 articles

Models

Articles

  • Alibaba Releases Qwen 3.8 27B with Native Vision and Dynamic Reasoning Controls

    Alibaba's Qwen research team has released Qwen 3.8 27B, an open-weight multimodal foundation model released under the Apache 2.0 license. The model combines 27 billion parameters with native vision processing, a 262,144-token maximum context window, and configurable inference-time reasoning controls. Under standard 4-bit quantization (Q4_K_M), the model compresses to approximately 17GB on disk, allowing local execution on consumer hardware with 24GB of VRAM or Apple Silicon unified memory syste

    1 min

  • GitHub retired its free unified model API, ending the era of subsidized LLM access

    GitHub pulled the plug on GitHub Models on July 30, 2026. The product was an odd but useful shape. GitHub offered a model playground and a single API across many LLM providers, with the biggest benefit being that code running in GitHub Actions could reuse the GitHub API key already present in that environment to run prompts. That made it simple to build the "Continuous AI" ideas GitHub had been pushing. It was also free or subsidized for developers. GitHub has not explained why it shut the ser

    1 min

  • ByteDance Launches SeedRealtime, a Full-Duplex Audio-Visual LLM

    ByteDance's Seed research team released SeedRealtime on August 5, a native audio-visual full-duplex large language model that fuses sound, vision, and text within a single unified architecture. Unlike cascaded systems that chain separate modules for speech recognition, vision, and text-to-speech, SeedRealtime runs perception, understanding, and response generation in parallel over continuous multimodal streams. The model has already been deployed in ByteDance's Douyin and Doubao consumer apps,

    1 min

  • Tencent Opens Hy3 to Global Users, Claims Top Spot on OpenRouter Within a Week

    Tencent announced global availability of its Hy3 large language model on August 5, expanding access beyond China through three channels: the WorkBuddy AI workspace, the Miora creative studio, and the Tencent Cloud TokenHub model-as-a-service platform. The rollout follows Hy3's initial release on July 6 and comes with a free access period on WorkBuddy through August 31. Hy3 uses a hybrid fast-and-slow-thinking Mixture-of-Experts architecture with 295 billion total parameters and 21 billion activ

    1 min

  • Mistral launches Shieldstral, a 3B open-weights policy-adaptive safety classifier

    Mistral AI released Shieldstral on Monday, a 3-billion-parameter open-weights safety classifier that matches or outperforms guard models up to seven times its size. The model is available under Apache 2.0 and runs on a single 16 GB GPU. What sets Shieldstral apart from typical guardrail models is its approach to content moderation. Rather than baking a fixed taxonomy of harm categories into the model weights — which forces developers to retrain whenever their safety requirements change — Shield

    1 min

  • Liquid AI releases LFM2.5-2.6B, a 2.6B model that rivals 4x larger models on agent tasks

    Liquid AI released LFM2.5-2.6B on Monday, a 2.6-billion-parameter model designed to run capable AI agents entirely on consumer hardware. The model is available on Hugging Face with open weights and targets on-device deployment across laptops and phones. LFM2.5-2.6B achieves 220 tokens per second on an Apple M5 Max and 113 tokens per second on an AMD Ryzen CPU, fitting within 2.5 GB of memory. Liquid AI positions it as competitive with models roughly four times its size on tool use, instruction

    1 min

  • Fable 5 comes back on a shorter leash

    Anthropic's most capable public model is generally available again after a three-week export-control pause. The interesting part isn't the model — it's the classifier stack now wrapped around it.

    1 min

  • Anthropic's enterprise quiet quarter

    While the discourse focused on consumer chatbots and benchmark drama, the real story was hundreds of regulated workloads quietly migrating to Claude. Here's why.

    1 min