#Open-source ecosystem

Meta releases Muse Spark Open, an 816B open-weights model

NVIDIA AI congratulates @alexandr_wang and the @Muse team on their open-weights model release. https://t.co/7JH7PXOnAz In the quoted post, @ArtificialAnlys reports that Meta's Muse Spark Open has 816B parameters (42B active) and scores 39 on the Artificial Analysis Intelligence Index, ranking it the #1 US open-weights model. Muse Spark Open (xhigh) scores 22 points above Meta's 30B dense Muse Glimmer (high, 17). https://t.co/jf31mQNINe

The same story is shown in the picks as 《Meta releases Muse Spark Open, scoring 39 on Artificial Analysis Intelligence Index》

China proposes export controls on open-weight models and 23 key AI technologies

China's commerce and science ministries released draft export restrictions targeting 23 key AI technologies, including large models, agents, compute clusters, and data processing. Controls on core model technologies would apply at 10^26 FLOPs of training compute, at least 2 trillion parameters, or global top-10 status, with separate restrictions covering models trained on China's important data. Open-source publication would qualify for automatic licensing, while installation on local servers abroad would require a standard export license; API access would remain exempt.

China proposes export controls on large models and 23 key AI technologies

China's commerce and science ministries released draft export controls covering 23 key AI technologies, including large models, agents, compute clusters, and data processing. Controls on core model technologies would apply at 10^26 FLOPs of training compute, at least 2 trillion parameters, or a global top-10 ranking. Open-source publication would qualify for automatic licensing, while installations on local servers abroad would require a standard export license; access through APIs, apps, websites, SaaS tools, and lightweight on-device versions would remain exempt.

Paradigm and Ruisizhilian Launch ModelSphere, an Open-Source AI Deployment Platform

Paradigm and Beijing Ruisizhilian announced ModelSphere, a free, open-source platform combining enterprise compute management, model deployment, and runtime optimization. Released under Apache 2.0, it supports heterogeneous chip scheduling and cluster operations alongside inference engines such as vLLM and SGLang, including cache-aware routing, autoscaling, and automated performance tuning. The company-provided announcement links to the repository at https://github.com/modelsphere/modelsphere.

Reflection-Beam-1.0 gets Day-0 support in vLLM

Reflection-Beam-1.0 from @reflection_ai has Day-0 support in vLLM. 501B total parameters, 23B active per token, and a 1M-token context window. Beam combines hybrid attention with Harmony-format reasoning and tool use. Beam also has Day-0 RL support in SkyRL (@NovaSkyAI): https://t.co/3SjWJKKgpH https://t.co/EkmA1L6B3H

Jev is Vercel’s fastest-adopted AI model ever

Jev reached 35% of paid teams within two weeks of launch, the fastest adoption recorded by Vercel and nearly twice the reach of Sonnet 5.5. Typesafe AI’s lightweight decision model selects among predetermined outputs for cheaper, faster classification. The figures come from Vercel’s AI gateway, which may undercount providers with their own APIs; open-source models now represent more than 70% of gateway tokens and nearly 25% of spending.

LIVE STREAM: New Open Weight Model https://x.com/i/broadcasts/1DxleVaPLagKL

LIVE STREAM: New Open Weight Model https://x.com/i/broadcasts/1DxleVaPLagKL

Reflection releases Beam, a 501B open model for coding and agentic work

Reflection released Beam on Hugging Face, a 501B MoE model with 23B active parameters per token, a 1M-token context window, and an Apache 2.0 license. The text-only model targets coding and agentic work, with BF16, FP8, and NVFP4 checkpoints and day-0 support in transformers, vLLM, and SGLang. Hugging Face provides deployment and post-training recipes alongside its own task evaluations and demos.

Reflection’s Beam scores 30 on Artificial Analysis Intelligence Index, ranking #2 among US open weights models

Reflection’s first model, Beam, scores 30 on the Artificial Analysis Intelligence Index, placing it #2 among US open weights models. Beam is an open weights model with 501B total and 23B active parameters, released on Hugging Face.

Reflection launches Beam, a 501B open-weight MoE model

Reflection launched Beam, a 501B MoE model, with open weights under the Apache 2.0 license and 40+ ecosystem partners. NVIDIA AI says it was trained on NVIDIA GB300 NVL72 and delivers over 70K tokens/sec/GPU on Blackwell Ultra out of the box on day 0. Reflection’s quoted announcement includes BF16, FP8 and NVFP4 weights, a DFlash2 drafter, and Mirror CLI; the Beam 1.0 NVFP4 checkpoint is available at https://huggingface.co/reflection/Beam-1.0-501B-A23B-NVFP4.

Hugging Face announces livestream on how Reflection built Beam

Hugging Face says a livestream with the Reflection team about how they built Beam starts in minutes. Stream: https://t.co/q6OKvCiB9o https://t.co/scNsrfsC3s In the quoted announcement, Reflection says Beam is available across platforms with open weights under Apache 2.0, in collaboration with 40+ partners, including a tokenizer, reference implementation, BF16, FP8 and NVFP4 weights, and a DFlash2 drafter.

Tencent releases Hy Memory 2.0 with an open-source model, framework, and MemHub

Tencent released Hy Memory 2.0 with an open-source model, framework, and MemHub, described as a private memory steward. The framework handles chat, work, and code memories. Tencent reports #1 open-source performance on 4 benchmarks and says Hy-Memory-30B-A3B beats Opus 5 across major frameworks. MemHub (CLI + Mac) syncs across 10+ AI tools. SDK & MemHub: https://github.com/Tencent-Hunyuan/Hy-Memory Model: https://huggingface.co/tencent/Hy-Memory-30B-A3B

NVIDIA and AI4Bharat Introduce Nemotron-pretraining-Indic-v1 and an Open Data Generation Pipeline

NVIDIA and AI4Bharat at IIT Madras introduce Nemotron-pretraining-Indic-v1, approximately 2.8 trillion tokens across 22 Indian languages, releasing soon alongside its open, GPU-accelerated generation pipeline. The corpus includes native-script text for all 22 languages and romanized text for twelve, translated from curated English sources with quality checks and source tracing. An Indic continued-pretraining stage improved five reported benchmarks, but combined synthetic and natural data, so the comparison does not isolate this corpus’s contribution.

Alexandr Wang Reacts to @kirillk_web3’s Muse Spark Post

@kirillk_web3 says Meta’s Llama 4 disappoints everyone, Meta goes closed with Muse Spark, Zuck writes a $100-200 billion check, Muse Wonder ships on Monday, and 3 days later Meta drops you with open weights: 816 billion parameters, only 42 billion active at a time. https://t.co/031gtR0nHm https://t.co/nEN1JMzvBa Alexandr Wang: “lmao muse spark open balling out” https://t.co/AcUT9o8NEQ

BFL releases FLUX 3 Image Klein with open weights and 5.6B parameters

BFL announced FLUX 3 Image Klein with open weights, 5.6B parameters, native 4K, bounding boxes, and pixel-perfect editing, saying it runs on a single consumer GPU. https://t.co/uIOfPdNhuL NVIDIA AI says the model was trained on NVIDIA GB200 NVL72, can run on an RTX GPU, and also has an NVFP4 variant: https://t.co/7rlalk7Otf https://t.co/zf1f7e4kJ7

The same story is shown in the picks as 《Black Forest Labs releases FLUX 3 Image Klein with 8 GB VRAM support》

Meta launches Muse Spark Open, an open-weight AI model designed for Macs

Meta has launched Muse Spark Open, a natively multimodal open-weight model with 816 billion parameters and a 1 million token context window, built for complex coding tasks and agentic workflows. It is designed to run locally on Macs, though Engadget reports it requires at least 1 terabyte of RAM. Meta claims superior performance over other popular open-weight models in its evaluations and is making the model available now through multiple API platforms.

The same story is shown in the picks as 《Meta releases Muse Spark Open, an open-weight multimodal mixture-of-experts model》

Token Rhythm Says Its Developer Base Has Passed 300,000 and Announces Double 11 Perks

Token Rhythm says more than 300,000 developers have joined its platform and has launched limited-time Double 11 presale offers. A single recharge of at least 50 yuan on TokenRhythm API earns an equal amount in model usage credits, with no cap on cumulative bonus credits; the company says the credits, including previously granted credits, will not expire. The article was provided by Token Rhythm and republished by QbitAI with authorization.

Ant open-sources Ling-3.1-flash for sustained work across tools and documents

Ant has open-sourced Ling-3.1-flash, a model with approximately 560B total parameters, about 25B activated per token, and a context window of up to 1M. QbitAI tested it on an e-commerce promotion plan and a simulated medical appointment preparation task, reporting traceable calculations and explicit handling of missing information. The review also found occasional attention drift with long inputs and noted that the current version lacks native visual understanding.

Meta releases Muse Spark Open for coding and agentic tasks

Meta has released Muse Spark Open, an open-weight model with 816 billion parameters (42 billion active), up to one million tokens of context, and text, image, video, and audio support. Meta reports leads over Qwen 3.8, GLM 5.3, and Kimi K3 in several benchmark categories, while trailing some competitors in coding. Weights are on Hugging Face; use requires the label "Built with Muse Spark Open," and API providers earning more than $20 million annually need a separate agreement with Meta.

The same story is shown in the picks as 《Meta releases Muse Spark Open, an open-weight multimodal mixture-of-experts model》

Guys this is like my superbowl I waited so long for US open weight models to be relevant again. Before today the US models were still behind Chinese models from as early as Feb. 2026. Still more work to do!

Guys this is like my superbowl I waited so long for US open weight models to be relevant again. Before today the US models were still behind Chinese models from as early as Feb. 2026. Still more work to do!

Guys this is like my superbowl I waited so long for US open weight models to be relevant again. Before today the US models were still behind Chinese models from as early as Feb. 2026. Still more work to do! This cut the gap by at least half, in time (say 3 months)

Guys this is like my superbowl I waited so long for US open weight models to be relevant again. Before today the US models were still behind Chinese models from as early as Feb. 2026. Still more work to do! This cut the gap by at least half, in time (say 3 months)

Clément Delangue Runs Muse-Spark-Open-816B-A42B-GGUF on Two Mac Studios

llama serve -hf ggml-org/Muse-Spark-Open-816B-A42B-GGUF:Q2_K (on two Mac studios). Well done @finkd & @johnternus!

Meta’s Muse Spark Open now available in Amazon SageMaker JumpStart

Meta’s coding-optimized open-weight model Muse Spark Open is now available in Amazon SageMaker JumpStart in BF16 and NVFP4 variants, with a 1 million-token context window and 816B total parameters, including 42B active per token. AWS provides deployment instructions using SageMaker Studio or the SageMaker Python SDK, plus inference and cleanup examples. At launch, availability covers US East (N. Virginia) and US East (Ohio).

Alexandr Wang Says Muse Spark Open Weights Are Out Today

Muse Spark Open — weights out today! Best US open-source model, and runs on the new Mac Studio 🍎 https://t.co/C8PU0RvlUA

Nathan Lambert Says He Needs to Blog After @AIatMeta Announces Muse Spark Open

@AIatMeta says it is releasing Muse Spark Open, which it calls “the most capable US open-weight model to date.” The quoted post says the model is built for coding and agentic work and lists 816B total and 42B active parameters, plus a 1M token [context window](https://t.co/yAtoxNeYJm). Nathan Lambert reacts: “Okay now I really gotta blog 🎉” https://t.co/uKl5X18Xpw

Nathan Lambert predicts strong muse spark open adoption if reported scores hold up

Nathan Lambert predicts adoption of muse spark open will surge if its performance roughly matches its reported scores, calling it the first US open-weight model with near-frontier agentic capabilities. He compares the shift with GLM 5.2 and Kimi K3 in China. In the quoted announcement, @alexandr_wang describes muse spark open as having 816B total params, 42B active, 1M-token context, and native multimodality, built for coding and agentic work. https://t.co/9elY7xigU5

Nathan Lambert Calls Meta Muse Open License “Trash,” Says Inference Providers Will Comply

The Meta Muse Open License is trash. Thankfully no one really cares about licenses, expect the inference providers who will happily comply.

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.

Black Forest Labs releases FLUX 3 Image Klein with 8 GB VRAM support

Black Forest Labs has released FLUX 3 Image Klein, a 5.6-billion-parameter open-weights model for text-to-image generation, image editing, and generation with up to ten reference images. It supports native 4K output, and its INT8 build runs on NVIDIA GPUs with 8 GB of VRAM, generating an image in under 9 seconds on an RTX 4070. The FLUX Kommunity License allows free commercial use for companies making under $5 million annually.

RedHat_AI releases optimized NVIDIA Nemotron 3.5 Super VL checkpoints

vLLM highlights @RedHat_AI’s FP8 and DSpark checkpoints for NVIDIA Nemotron 3.5 Super VL, a 120B multimodal MoE, using LLM Compressor and Speculators. RedHat_AI reports close to 2x performance on Hopper with accuracy intact. Its 4× H100 benchmarks show FP8 improving throughput under load and DSpark boosting per-user streaming speed, with both together outperforming either alone across HumanEval, math reasoning, and RAG workloads. Both checkpoints are open: https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-FP8 https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B.dspark

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.

Nvidia releases Nemotron 3.5 Super VL for continuously running enterprise AI agents

Nvidia released Nemotron 3.5 Super VL, an open multimodal model with 120 billion parameters and 12 billion active per token, for continuously running enterprise agents. Nvidia says it approaches Nemotron 3 Ultra’s accuracy on specified agent tasks with about one-fifth the parameters, and is releasing weights, datasets, and training recipes. The accompanying Nemotron 3.5 Bolt 74B-A9B 24GB targets local agents, while Bolt E4B 4GB is planned for later this month.

vLLM-Omni adds day-0 support for FLUX 3 Image Klein

vLLM-Omni now supports FLUX 3 Image Klein from day 0. The vLLM team worked with Black Forest Labs and NVIDIA to bring the new image model to its serving stack. The integration includes FLUX 3 Image Klein image generation and BF16, FP8 and NVFP4 checkpoints. https://t.co/hsjdzMrbkt

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 Expands Nemotron Family With Super VL and Bolt Open Models

NVIDIA released Nemotron 3.5 Super VL, an open 120-billion-parameter model for multimodal agents, and Nemotron 3.5 Bolt 74B-A9B 24GB for local agentic workloads; Bolt E4B 4GB is coming Oct. 23. Super VL targets tool use, business workflows and visual document understanding, with open weights, datasets and training recipes. Bolt 74B-A9B uses native 2-bit quantization to fit in 24GB of memory, while Bolt E4B targets physical and conversational agents.

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

Apertus 1.5 Release Notes Detail Multimodal Understanding, Thinking Mode, and Longer Context

Project Apertus released Apertus 1.5 in 8B and 70B sizes, adding image and audio understanding, optional thinking, a 262,144-token context window, and improved instruction-following and tool use. The weights were released on Hugging Face on July 24, 2026 under Apache 2.0; the models generate only text, and spoken-language processing remains experimental. The team also publishes training data, recipes, intermediate checkpoints, and reproducible evaluation records.

vLLM adds day-one support for Jais 3

Day-one support for Jais 3 is available in vLLM. Congratulations to @IFM_AI and collaborators on the launch. Arabic-native agentic AI, ready to serve with vLLM from day one. https://twitter.com/IFM_AI/status/2110641827778843010

NVIDIA Expands Tools, Training and Compute Access for Europe’s Scientific Community

NVIDIA is joining the European Commission’s AI in Science Enablers Initiative, pledging tools, training, compute access and collaborations for Europe’s research community. Its Deep Learning Institute has launched freely downloadable, openly licensed course materials with no enrollment, fees or deadline. NVIDIA is also collaborating with European AI factories to expand compute access for startups and SMEs, while 35 supercomputers under development across Europe are intended to serve over 3 million researchers.

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.