Meta releases Muse Spark Open, an open-weight multimodal mixture-of-experts model
Meta’s Muse Spark Open is an open-weight mixture-of-experts model with approximately 816 billion parameters, around 42 billion active per token, and a context window of up to one million tokens. Trained for agentic coding, it accepts text, images, audio, and video and generates text. Hugging Face provides inference and deployment recipes, reports its own quantization tests, and demonstrates attention-layer LoRA fine-tuning with TRL while keeping the experts frozen.
Model releases and updates
NVIDIA Releases Nemotron 3.5 Super VL and Bolt Models for Server and Edge Agents
NVIDIA released Nemotron 3.5 Super VL, an open 120-billion-parameter multimodal model, and Nemotron 3.5 Bolt 74B-A9B 24GB, with Bolt E4B 4GB coming Oct. 23. Super VL targets specialized agent workflows involving tools, documents, and visual understanding, while Bolt models target local and edge workloads. NVIDIA provides open weights, datasets, and training recipes for Super VL; Bolt 74B-A9B uses native 2-bit quantization to fit in 24GB of memory.
Kunlun Tech releases and open-sources Matrix-Game 4.0 for single-GPU real-time gameplay
Kunlun Tech released Matrix-Game 4.0, reporting 480p gameplay at 60 FPS on a single NVIDIA RTX 5090 with end-to-end response latency below 150 milliseconds. Its game engine maintains world state and interaction rules, a coding agent modifies game logic, and a video model generates detailed visuals from structural frames and dynamic semantic conditions. Code, weights, and simplified scene models are available at https://huggingface.co/Skywork/Matrix-Game-4.0.
Mistral AI releases Timestral, a family of time-series foundation models
Mistral AI released Timestral in four sizes, from 5M to 209M parameters, for probabilistic, multivariate forecasting with past-only and known-future covariates. Mistral reports that its 209M model ranks first among pretrained, zero-shot models across five metrics on GIFT-Eval, TIME and fev-bench, using a single checkpoint without ensembling. All sizes support CPU inference, and Mistral published a GitHub repository for running Timestral on users’ data; domain adaptation and fine-tuning are available through Mistral Forge.
StepFun releases open weights for Step 5 Preview, a 600B MoE model
NVIDIA AI congratulates StepFun on Step 5 Preview, calling it StepFun’s largest open weights release. NVIDIA says the 600B MoE model supports agent tasks lasting up to 24 hours and runs at nearly 8K tokens/sec/GPU on NVIDIA GB300 using NVFP4. StepFun’s quoted announcement lists 27B active parameters per token, a 1M-token context window, native text and vision input, and BF16 and FP8 weights on Hugging Face and ModelScope. NVP4 checkpoint: https://t.co/lSECChAwlW Deploy with https://t.co/GO1wqTSBOk https://t.co/28TBzUo6ZU
MAI announces in-house pre-trained base models in the 3T-parameter weight class
We now have state-of-the-art pre-trained base models in the 3T-parameter weight class, built entirely in-house at MAI. Post-training results coming soon—stay tuned! Lots of details in our technical blog post. https://microsoft.ai/news/frontier-pre-training-at-microsoft-ai/
Product releases and updates
Welcome FLUX.3 Image Klein in Diffusers
Black Forest Labs’ FLUX.3 Image Klein launches with day-one Diffusers support for text-to-image generation, multi-reference image-to-image generation, and bounding-box layout control. Open checkpoints include a distilled model using 16 steps without CFG and a base model using 50 steps with CFG; the distilled model generated a 1024 × 1024 image in 1.93 s on an H100 in the reported benchmark. Hugging Face provides quantized loading examples, LoRA training scripts, and workflows for customizing its Modular Diffusers pipeline.
NVIDIA DLSS 4.5 Controls Arrive on GeForce NOW
NVIDIA has added DLSS 4.5 controls to GeForce NOW, letting members adjust image quality and performance in supported games. The settings are available in the app on Windows, macOS, Chrome OS and Steam OS, with Ultimate members on GeForce RTX 5080-powered cloud rigs gaining options including Dynamic Multi Frame Generation up to 4X. Enshrouded 1.0 and Warhammer 40,000: Boltgun 2 are also streaming at launch, alongside 10 games joining the service this week.
Industry
The Verge · AIEiliko robot toy exposes private conversations and allows remote manipulation, The Verge reports
Sean Hollister tested Energize Lab’s Eiliko robot toy with hacker Sammy Azdoufal, who accessed his conversation transcript and remotely altered the toy’s memory and responses. Azdoufal reported access to 13 million conversations and seemingly the entire active fleet of 68,000 toys, alongside detailed user profiles and publicly accessible photos and audio recordings. Energize Lab did not answer their security questions despite repeated outreach.
Research
Shanghai AI Laboratory’s Shusheng-S2 Helps Prove the Three-Dimensional Ghys Conjecture
Xu Disheng’s team used Shanghai AI Laboratory’s Shusheng-S2 model in a human–AI collaboration to prove the three-dimensional Ghys conjecture, with a preprint posted on arXiv. Researchers guided seven rounds of recursive exploration as the problem graph expanded from 15 to 389 nodes. A Peking University team used its SHEAF framework to formalize all four main theorems, producing approximately 550,000 lines of Lean code in about five days and passing Lean checks.
Microsoft Research presents Magnetar for proof-based general-purpose computing
Microsoft Research presents Magnetar, a zero-knowledge virtual machine that proves RISC-V program execution using security based only on standard hash functions. It supports programs written in Rust, Go, or C without requiring developers to rewrite them as cryptographic circuits. Across 15 benchmark programs, Microsoft reports average speeds 10 to 39 times faster than the fastest existing comparable zkVMs on the same machine, with proofs verified in about 12 milliseconds.
The DecoderGPT-6-Astra solves 49 of 50 physics research problems at $2.40 apiece
GPT-6-Astra, running in OpenAI's Codex environment without human hints, solved 49 of 50 TheoPhysEval problems across eleven theoretical physics subfields, costing just under $116 overall. Anthropic's Opus 5.5 scored 92 percent, while both models reached 92 percent without internet access. The researchers emphasize that the benchmark tests calculations using known methods, while choosing promising research questions remains a separate challenge for current agents.