Meta Releases Muse Spark Open Weights on Hugging Face
Meta released the weights for Muse Spark Open, described as the most capable US open-weights model to date. The model is available today on Hugging Face under the Muse Model License, according to a PyTorch blog post announcing the release.
Sources and evidence
Attributed quotes
“Today, Meta is releasing the weights for Muse Spark Open, the most capable US open-weights model to date. It's available today on Hugging Face under the Muse Model License.”
Summary last validated Oct 17, 2026
Reader actions
Report an issue
Use this for an incorrect summary, wrong source, duplicate story, or wrong category. Submissions are private and do not change the story automatically.
Related coverage
- models
Amazon Bedrock adds structured outputs for Anthropic Claude models
Amazon Bedrock now supports structured outputs for all Anthropic Claude models on both the bedrock-runtime and bedrock-mantle endpoints. The feature returns schema-compliant JSON matching a defined structure using JSON Schema output format and strict tool use, eliminating the need for custom parsing, retry logic, or post-processing.
- models
Cloudflare introduces Clairic System One extraction models
Cloudflare announced Clairic, a new model it describes as a System One extraction model. According to Cloudflare, Clairic can extract, compute, estimate, and forecast data from unstructured inputs. The company detailed the model in a blog post introducing the new class of extraction models.
- models
TwelveLabs Pegasus 1.5 video model launches in Amazon Bedrock
TwelveLabs' Pegasus 1.5 is now available in Amazon Bedrock. The video language model converts long-form video into structured, timestamped JSON metadata by matching user-defined moments. New features include video segmentation, multimodal prompting with reference images, and on-the-fly processing of up to two hours of video.
- open source
PyTorch Helion integrates NVIDIA CompileIQ for compiler controls
PyTorch's blog describes how Helion, its tool for writing GPU kernels in PyTorch-native Python with an autotuner, now incorporates NVIDIA CompileIQ to add compiler controls. The post frames this as a two-layer search approach, pairing Helion's autotuning with CompileIQ's compiler-level controls.
- infrastructure
PyTorch Introduces TorchNeuron Native Backend for AWS Trainium
PyTorch announced TorchNeuron, a native PyTorch backend for AWS Trainium. The blog post says frontier models now span language, video and audio across dense and mixture-of-experts transformers, diffusion, state-space, world and structure-prediction architectures, and frames PyTorch as where AI research and models are built. The post provides no further technical details in the excerpt.