Latest

A sparse 12B-active multimodal model running free inside a coding agent changes the cost math for developers running agentic coding sessions at volume. Engineers can point Cline at Nemotron 3.5 Super VL today and compare its video and document handling against Nemotron 3 Ultra directly.

Cline@clineSource

The distribution breadth is the notable part: day-one hosting across seven inference providers plus four weight mirrors makes it easy to test Muse Spark Open without Meta's own stack. The Muse Model License's limited commercial terms mean teams should check whether their revenue scale qualifies for free use before shipping to production. Developers comparing open-weight options can benchmark it directly against similar offerings on Groq or Together AI.

Official accountMeta AI@AIatMetaSource

A 42B active parameter count suggests inference costs far below what the 816B total implies, which matters for teams self-hosting open weights. The 1M context window and native audio and video input position it as a direct alternative to closed frontier models for agent pipelines. Developers building on open-source models now have a US-trained option to weigh against Chinese open-weight releases like DeepSeek and Qwen.

Official accountMeta AI@AIatMetaSource

Combining vision and text in one model lowers latency and engineering overhead for agent pipelines that previously chained a vision model behind an LLM. Developers building document-reading, chart-interpreting agents should evaluate whether Nemotron 3.5 Super VL can replace their two-model setup and simplify context management.

Official accountNVIDIA AI@NVIDIAAISource

NVFP4 quantization on a 12B-active MoE points to a price point well below most multimodal API options, making cheap vision-plus-tool-calling agents practical to deploy. Builders prototyping agentic workflows can test it on CoreWeave's serverless endpoint and compare its per-million-token cost against frontier multimodal APIs before committing. Traceability through Agent Lens also matters for teams debugging structured-output pipelines.

CoreWeave@CoreWeaveSource

Speed, not raw accuracy, is the headline here: a 55% reduction in total task time against DeepSeek V4.1 Flash suggests latency-sensitive agent pipelines are where this model fits best. Developers building Deep Agents workflows in Python can test it directly through ChatNVIDIA with open weights, making self-hosting or cost comparison against closed APIs straightforward.

Official accountLangChain@LangChainAISource

Day-0 support means developers can self-host the model with vLLM immediately instead of waiting on inference framework integration, and the sparse 12B-active MoE design keeps serving costs well below a dense 120B model. Builders running agents over long documents or video pipelines can start with the official vLLM setup guide in the announcement thread.

Official accountvLLM@vllm_projectSource

Releasing data and recipes alongside weights lets practitioners reproduce or fine-tune the full training pipeline rather than just the checkpoint, which is rare even among open model releases. Builders running multimodal workloads can compare it against other open vision-language models and adapt it for their own inference setups.

Official accountNVIDIA AI@NVIDIAAISource

A sparse 120B architecture with only 12B active parameters suggests NVIDIA is optimising for agent workloads where inference cost matters more than raw size, and the open weights let teams self-host rather than depend on frontier API providers. Developers building document- or video-processing pipelines should evaluate it against Qwen-VL and other open multimodal models for agent deployment.

Official accountNVIDIA AI@NVIDIAAISource

Specialisation, not scale, is the pitch: NVIDIA supplies open post-training tooling while Scale AI contributes the data and evaluation layer, so the joint value is a pipeline for tailoring Nemotron 3.5 Super VL to vertical workloads. Enterprises weighing specialised fine-tunes against frontier general models, and teams already in the NVIDIA ecosystem, are the natural audience. Availability details and pricing were not in the announcement.

Official accountScale AI@scale_AISource

The interesting signal is the delivery model: a lab embedding researchers with customers to fine-tune an open NVIDIA base model cheaply, instead of selling API access to a proprietary frontier model. Engineering and AI leads evaluating build-versus-buy for domain models can use the thread's results as a reference point. Deel's involvement gives it a hiring-and-HR use-case angle.

Official accountAI21 Labs@AI21LabsSource

Together AI hosting means the open weights can be served without NVIDIA's own inference stack, giving teams more deployment options for agentic vision workloads. Developers building document-processing or screen-automation agents can compare it with closed multimodal APIs on cost and control before committing.

Official accountTogether AI@togethercomputeSource

The claim of 24-hour agent endurance points at long-horizon automation rather than chat benchmarks, and the near-8K tokens/sec/GPU throughput figure shows quantised inference on GB300 is becoming a default release channel. Builders deploying open-weights agents can start from the SGLang or vLLM recipes. StepFun's open release also gives ASEAN teams an alternative to closed frontier APIs for self-hosted workloads.

Official accountNVIDIA AI@NVIDIAAISource

The gap between Google and the ECI frontier has narrowed to 2 points, putting Gemini back in competitive range after months of lag. Teams choosing models for hard reasoning or agentic workloads now have a four-way race to benchmark on their own tasks rather than defaulting to a single leader. Comparable scores across Opus 5.5, GPT-6 Astra and Argon suggest provider switching costs are lower than vendor benchmarks imply.

Official accountEpoch AI@EpochAIResearchSource