The Information: DeepSeek launches V4-Pro, its most advanced model that rivals Kimi K3 on some benchmarks at much lower prices, costing $0.44/1M input and $0.87/1M output tokens — Chinese AI developer DeepSeek…
via theinformation.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
The lets you run established coding-agent runtimes through one unified interface, so you can switch runtimes without changing your application code. Today we are adding Grok Build, which runs through the same interface…
via vercel.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
The now supports any Agent Client Protocol (ACP)-compatible harness with through the new package.AI SDK harness layerHarnessAgent@ai-sdk/harness-acp Previously, every harness adapter wrapped one specific runtime…
via vercel.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
xAI's Grok 4.6 scores 61 points on the Artificial Analysis Intelligence Index, tying GPT-5.6 Sol and trailing only Anthropic's Claude Opus 5. On agentic tasks, it completes complex workflows in about 53 steps where…
3.0.0 (2026-08-12) ⚠ BREAKING CHANGES api: HTTPX2 is now the default HTTP client, and `httpx` is no longer installed automatically. Applications using custom HTTPX clients, transports, or configuration objects must…
via github.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
Mistral AI Blog: Mistral says its platform will support third-party open models, starting with Z.ai's GLM-5.2, and run them on the same infrastructure as its own models — At Mistral, we believe every enterprise…
via mistral.ai·Click to report a broken or paywalled link. Two distinct reports hide the row.
DeepSeek V4 Pro now runs on updated weights on AI Gateway. They are used by default when you call , so existing requests pick them up with no change to the model ID or your code.deepseek/deepseek-v4-pro To use the…
via vercel.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
Using coding agents means setting up multiple accounts, provisioning API keys, and scattering observability and billing. Now, you can route them through to centralize all of this and add controls, with set up in one…
via vercel.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
Vercel Connect now gives teams line-level visibility into the token lifecycle. See who created a token, what app or project used it, when, and whether it is still active. Every connector's detail page includes a new…
via vercel.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
NVIDIA released Nemotron 3.5 Lightning, a 30B mixture-of-experts model with 3B active parameters for agentic tasks, alongside NeMo Switchyard, an open-source library for model routing in agent workflows.
via marktechpost.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
LTX launched LTX-2.5, an NVIDIA-accelerated open-weights world model for video generation, alongside NVIDIA's release of the 30B Nemotron 3.5 Lightning agent model.
via marktechpost.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
Accelerate cyber defense with OpenAI and AWS: Daybreak Red & Daybreak Blue now available to eligible customers on Amazon Bedrock Amazon Web Services (AWS)
via Google News — AI·Click to report a broken or paywalled link. Two distinct reports hide the row.
What's Changed Add Muse Glimmer support for NVIDIA, AMD, and additional platforms Full Changelog: https://github.com/ollama/ollama/compare/v0.32.7...v0.32.8-rc0
via github.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
Meta has released Muse Glimmer, the first open model from its new Superintelligence Labs. It's a 30B agent model that runs on consumer hardware once the weights are compressed, needing less than 20 GB of memory. In an…
via the-decoder.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
vLLM v0.27.0 Release Notes Highlights This release features 561 commits from 242 contributors (64 new)! Kimi K3 support with a full stack landing in one release: core model files and kernels (#50089, #50000), Python…
via github.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
Hey HN,Henry from Cactus here!We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got…
via cactuscompute.com·Click to report a broken or paywalled link. Two distinct reports hide the row.
NVIDIA released NemotronLabs VoiceChat 11B, an open 11B end-to-end speech-to-speech model capable of full-duplex conversation with ~450 ms latency and live tool calling.
via marktechpost.com·Click to report a broken or paywalled link. Two distinct reports hide the row.