iCreatAI News
Back to main site
Skip to content
#Open source
  1. Hacker News AI78

    GitSeed – Self-hosted control plane for AI coding agents

    GitSeed is an Apache-2.0, self-hosted, vendor-neutral control plane that runs coding agents such as Claude Code, Codex, Gemini, and OpenCode inside isolated three-phase containers (prepare → agent → finalize) on your own infrastructure. It adds Ed25519-signed PR Governance Receipts, parallel three-model "Second Opinions," enforced per-run spend caps via LiteLLM, and a five-stage SDLC pipeline (Ideate → Plan → Execute → Review → Merge) or Fast-Track mode, with a CLI/TUI, a 65-tool MCP server, and a Helm chart.

    Why it matters: Useful for teams weighing self-hosted agent orchestration against cloud SaaS bots, since it details concrete trust boundaries, audit receipts, and spend controls.

  1. The Decoder85

    Meituan releases a 1.6-trillion-parameter agent model that never touched an Nvidia GPU

    Meituan has released the weights for LongCat-2.5, a 1.6-trillion-parameter mixture-of-experts model for autonomous agent and coding tasks that activates roughly 48 billion parameters per token and was trained entirely on Chinese hardware. It handles up to one million tokens of text, image, and video input via API, and Meituan reports 88.4 on Terminal Bench v2.1 and 64.6 on DeepSWE, close to but below Claude Opus 5 and GPT-5.6 Sol on several benchmarks. The weights are MIT-licensed, and the self-hosting guide describes a 194-chip Huawei Ascend Atlas A2 cluster with no Nvidia GPU guidance.

    Why it matters: It documents a frontier-scale open-weight agent model trained and served entirely on Chinese accelerators, with benchmark comparisons against leading Western models.

  1. Cloudflare Blog71

    Kitesurf is now open source

    Cloudflare has open-sourced Kitesurf, its stateless browser for AI agents that runs entirely on Workers, under the Apache 2.0 license on GitHub. The repository includes the engine, PageScript, PageRenderer, and CLI, built on Rust-to-WebAssembly projects like Blitz, Stylo, Boa, and Linebender libraries, and supports the Chrome DevTools Protocol for clients such as Puppeteer. Developers can run it locally or deploy their own instance to Cloudflare Workers, which requires a Workers Paid plan because it uses Dynamic Workers.

    Why it matters: Useful for developers who want to self-host a lightweight agent-oriented browser engine and inspect or contribute to its open-source components.

  1. Hugging Face Blog60

    llama.cpp v1.0.0 : A stronger foundation for the local AI ecosystem

    llama.cpp released libraries v1.0.0, marking a commitment to stable public APIs for downstream developers, with both llama.cpp and ggml now reflecting API and ABI incompatibilities through major-version bumps. The project states it supports more than 150 architecture variants and all mainstream speculative decoding methods, and distributes pre-release artifacts across Linux, macOS, Windows, Android and iOS plus channels including Docker, Homebrew, conda-forge, MacPorts, Winget and XCFramework. It also describes continued investment in CI infrastructure and self-hosted runners with hardware support from NVIDIA, AMD, Qualcomm, Arm, IBM and Intel.

    Why it matters: Signals a versioning milestone for a widely depended-on local inference library, useful for developers tracking API stability and platform support.

  2. Hacker News AI88

    Edge0-145B: Frontier AI Capabilities on Your Own Computer

    The company open-sourced Edge0-145B and Edge0-120B, describing Edge0-145B as the first open-source 145-billion-parameter on-device model, alongside a unified inference framework and deployment tools. It reports 29GB peak memory, 80+ tokens/s decoding, 1,100+ tokens/s prefill, roughly 120W typical power and 128K–256K context on a compatible laptop, with benchmark scores compared against named cloud models. The release also describes shared-matrix sparse experts, tiered GPU/system-memory/SSD loading and CPU-GPU co-scheduling, plus local demonstrations in video, Blender 3D scenes, games, drone control and PCB design.

    Why it matters: It offers a detailed, benchmark-backed look at how a 100B+ model is engineered to run locally, useful for readers weighing on-device versus cloud AI.

  3. Hugging Face Blog66

    LLMON: An LLM-Native Markup Language, now in Granite 4.1 3B and 8B

    IBM researchers introduce LLMON (LLM Object Notation), an XML-like markup language that uses reserved special-token delimiters to explicitly mark instructions, data, requirements and execution bindings in an LLM's context, addressing the brittleness of JSON-style formats for causal models. They release two experimental preview models derived from Granite 4.1 base checkpoints (granite-4.1-3b-LLMON-preview and granite-4.1-8b-LLMON-preview) plus a GitHub toolkit, reporting that on the OPI benchmark LLMON models avoid 47% more indirect prompt injection attacks than the best frontier model tested while maintaining general capability on par with Granite 4.1 Instruct. The models ship with their own chat template and lossless JSON-to-LLMON converters, and the post notes the previews are for exploration, not production.

    Why it matters: Useful for teams weighing structured context formats against JSON and for anyone evaluating prompt-injection defenses in small open models.

  1. vLLM Blog63

    H3-Image Day-0 Support: 4K in 0.87s with Vera Rubin

    vLLM-Omni brought MiniMax H3-Image, an open-weight text-to-image and editing model, to NVIDIA Vera Rubin, reaching 0.87 seconds median for native 4096×4096 generation on four GPUs with SP4 and MXFP8 in a short-prompt serving benchmark. A separate SP1×DP4 throughput configuration delivers 6,115 images/hour, which the authors project to 146,755 images/day and, at an assumed USD 0.06879 per image, USD 105.16/GPU/hour gross revenue at full billable utilization before operating costs. The post also documents VAE decoding gains from CUDA Graph replay plus multiple streams (2.635 s to 0.213 s on four GB300 GPUs) and community validations on AMD, Intel, Huawei Ascend and Moore Threads hardware.

    Why it matters: Useful for readers weighing open-weight 4K image serving, since it separates latency and throughput operating points and labels its revenue projections as extrapolations before operating costs.

  1. AWS Machine Learning Blog60

    UAEF: A framework-agnostic system for closed-loop evaluation of AI agents

    AWS introduces the Universal Agent Evaluation Framework (UAEF), an open-source, framework-agnostic system that uses adapters to normalize agent traces from LangGraph, LangChain, Strands Agents, Amazon Bedrock AgentCore, and Langfuse into a common AgentTrace format, then scores them with 36 built-in metrics across 7 quality dimensions. Results are stored per experiment in Amazon DynamoDB and Amazon S3, enabling comparison of agent versions, regression detection, and report and insight generation, with both a Python library and a low-code web UI sharing the same engine. The post also provides a playbook for choosing metrics and preparing datasets, plus runnable notebooks for each supported framework.

    Why it matters: It offers a detailed, practical reference for teams evaluating agents across frameworks, including metric selection guidance and a closed-loop workflow for catching regressions before release.

  2. Hacker News AI78

    Show HN: AI solved software. Now it's time to raise the bar. Introducing Brigade

    Brigade is an MIT-licensed open-source product that gives you a team of AI teammates running the real Claude Code or Codex CLIs signed in with your own Claude or ChatGPT subscription accounts, each with a job, its own computer, Apps and memory. Decisions needing you land in one list on your phone, and it is free to self-host or $99 a month per workspace for Brigade Cloud. The repository includes the dashboard, API, runner, desktop app and deployment docs, and most commits come from Brigade's own teammates with a person approving every change.

    Why it matters: Useful for readers weighing agent orchestration tools, since it details self-hosting, subscription-based usage and approval-gated writes.

  3. PyTorch Blog71

    ExecuTorch WebGPU: PyTorch Models on the GPU, in the Browser

    ExecuTorch's new WebGPU backend runs PyTorch models exported with torch.export to a .pte file directly on the GPU in WebGPU-enabled browsers, without converting to another framework's format. The same artifact runs across Chrome and Safari on Metal, Vulkan and D3D12, and on an Apple M1 Pro in Chrome it was 4–7× faster than ExecuTorch's Wasm (XNNPACK) CPU path on prefill and 2–10× faster on decode. Validated models include Llama 3.2, Qwen3, Gemma 3, SmolVLM, Depth Anything V2, SAM 2, Whisper and on-device LoRA fine-tuning, all running locally in the tab.

    Why it matters: It documents a concrete path for running PyTorch models on browser GPUs with benchmark numbers and a list of validated models, useful for developers weighing in-browser inference.

  4. Amazon Science62

    Mitra-v2: Scaling synthetic pretraining for tabular foundation models

    Mitra-v2 is a second-generation tabular foundation model for classification and regression, pretrained entirely on synthetic tables with 10x as many rows and roughly 3x as many columns as Mitra-v1, plus a new hybrid SCM prior and hybrid optimizer. On TabArena it leads the single-model field, beating Mitra-v1 on 32 of 33 datasets where both run natively and completing all 51 datasets within the one-hour budget, and fine-tuned Mitra-v2 ranks first on ScoringBench. The fine-tuned models and inference/fine-tuning code are released under Apache-2.0 on Hugging Face.

    Why it matters: Useful for readers tracking tabular foundation models, as it reports concrete benchmark rankings and an open Apache-2.0 release.

  1. n8n Blog76

    Meet Nod: a standalone personal agent from n8n Labs that runs on your computer

    n8n Labs, a new team within n8n, has released Nod as a research preview: a standalone, open-source (Apache 2.0) personal agent for Mac and Windows that runs on your computer with its own local sandbox and browser, working in connected apps like Gmail, Calendar, Drive and Slack as well as websites with no integration. It remembers months of conversation (scoring 93.2% on LongMemEval), offers jobs it could take on, runs tasks in parallel via the app, Telegram or WhatsApp, and is free apart from model provider costs, with per-task receipts and a monthly spending limit.

    Why it matters: Useful for readers weighing a local, open-source personal agent, since it details Nod's sandboxed design, permission checks, memory benchmark and cost controls.

  1. NVIDIA Newsroom78

    Practice Makes Precise: At CoRL, NVIDIA and Partners Reveal How Robots Rehearse Delicate Work in Simulation

    At CoRL in Austin, NVIDIA introduced open robotics technologies including GR00T 2, an open world action model built on NVIDIA Cosmos and trained on diverse robot and human data including 100,000 hours of open source human egocentric data, plus Isaac Lab 3.0 with tighter Newton integration for deformable objects and a new reinforcement post-training library. Two NVIDIA papers, GRAIL and RoboTTT, are among six finalists for Best Paper Awards, and partners including Ambi Robotics, Bosch Research, Siemens, Skild AI and UC Berkeley are applying the tools to benchmarks such as DexBench and VABAR and to agentic workflows like Graph-as-Policy. The piece also details AlphaNIST, which trains a single policy across all 20 NIST Assembly Task Board tasks with over 98% training success using roughly 39x less GPU time, and Isaac Lab 3.0 with Newton reaching up to 2.5x the training throughput of Isaac Lab 2.3 with PhysX in a vision-based dexterous manipulation benchmark.

    Why it matters: It offers a detailed, source-grounded survey of NVIDIA's CoRL announcements and partner deployments, useful for readers tracking simulation-based robot learning and evaluation benchmarks.

  1. vLLM Blog72

    Zero Mismatch, Simple Async RL: REINFORCE Zero, DeepSeek-V4.1 and Kimi K3.1

    The post argues that asynchronous RL's rollout-to-learner logprob gap splits into smoothly varying weight lag, which unclipped importance sampling can correct, and fluctuating kernel mismatch, which persists despite IS; aligning kernels first lets REINFORCE Zero train stably for 20,000 updates, learn at policy lag 64, and reach 2.40× the training-step throughput of synchronous training with aligned kernels. It reports zero mismatch (bit-for-bit logprob agreement) for DeepSeek-V4.1-Flash and Kimi K3.1 on day 0, and for DeepSeek-V4-Flash and Kimi K3.1 on AMD MI355X. Code and recipes are in Inferact/zero-mismatch-rl, with the paper at hijkzzz/REINFORCE-Zero and training code in NVIDIA-NeMo/labs-molt.

    Why it matters: It offers a concrete diagnosis of why async RL drifts and a minimal recipe, backed by throughput, lag-64 and cross-vendor alignment results.

  2. Hugging Face Blog62

    New NVIDIA Cosmos 3 Open Models for Interactive World Simulation

    NVIDIA's Cosmos world model platform adds four new open-weight checkpoints to its Cosmos 3 family: Cosmos3-Nano-Transfer-Auto for multi-view driving data with varied weather, lighting and time of day; Cosmos3-Nano-Sim-Auto for streaming synchronized camera and LiDAR driving simulation; Cosmos3-Nano-Sim-Transfer for turning simulator controls and text prompts into photorealistic video; and Cosmos3-Nano-Sim-Bimanual for predicting video from robot or camera actions. All four are post-trained examples of Cosmos 3 Nano with 16B trainable parameters and use OpenMDW1.1, permitting commercial and non-commercial use under its terms. The post also details a four-stage training recipe (bidirectional adaptation, teacher forcing, consistency distillation, self-forcing) and post-training guidance for adapting the models to new sensor rigs and embodiments.

    Why it matters: It gives a concrete inventory of newly released open-weight world-model checkpoints plus the training recipe, useful for researchers weighing them for robotics or driving simulation.

  3. Artificial Analysis Articles62

    Nanbeige has released Nanbeige4.3-33B scoring 20 on the Intelligence Index

    Nanbeige, the foundation model research team at BOSS Zhipin (operated by Kanzhun Limited), has released Nanbeige4.3-33B, an Apache 2.0 open-weights dense reasoning model with text-only input and output and a 524K-token context window. It scores 20 on the Artificial Analysis Intelligence Index v5, matching Qwen3.8 27B (xhigh) at 20, with relative strengths in terminal use (30% on Terminal-Bench 4.0 vs. 6%) and physics reasoning (16% on CritPt v2's Hard Subset vs. 4%), but trails on AA-Briefcase v1.1 (1162 vs. 1391 Elo), FrontierCode 2.0 (11% vs. 15%) and HLE-Diamond (7% vs. 15%). It shows high abstention and low hallucination on Omniscience (18% vs. 29% hallucination rate, declining 71% of questions) and used 158K output tokens per Intelligence Index task vs. 110K for Qwen3.8 27B.

    Why it matters: Useful for readers tracking small open-weights models, since it sets a new Pareto point for Intelligence Index versus total parameters and details trade-offs against Qwen3.8 27B.

  1. Hugging Face Blog71

    A Closer Look at Phonon TTS. Launching High-Quality On-Device Speech That Tops the Benchmarks.

    Gradium details Phonon, a small on-device text-to-speech model released two weeks ago that needs no server or network, with Phonon 31m (English) under Apache 2.0 and Phonon 40m covering English, French, German, Spanish and Portuguese under a non-commercial license. The models, about 65 MB and 75 MB, offer over 100 built-in voices plus voice cloning and voice design, and on Hugging Face's Open TTS Leaderboard Phonon ranks first in German and Spanish on intelligibility and tops the CPU streaming benchmark on time-to-first audio. The post also covers the ptts inference framework, Transformers integration from v5.21, launch demos, and more than 200,000 downloads in the first two weeks.

    Why it matters: It offers a detailed look at how a small on-device TTS model was built, benchmarked and integrated into developer tooling.

  2. vLLM Blog72

    Introducing Kinference: An Open Source Inference Control Plane for vLLM and Kubernetes

    Buoyant announced Kinference, an Apache v2 open source control plane for building inference platforms on vLLM and Kubernetes, offering token rate limiting, tenant usage and cost tracking, KV-cache-aware load balancing, and a YAML-based inference policy engine. The post reports benchmarks in which Kinference cut median time-to-first-token by 95% and median end-to-end latency by 45% versus standard Kubernetes load balancing on one workload, and served 7.4× the goodput on a shared-prefix workload and about 3.5× on multiturn. It also covers per-tenant inference metrics and Helm-based installation steps.

    Why it matters: It gives a concrete look at cache-aware routing and policy-driven model selection for self-hosted vLLM deployments, backed by published benchmarks.

  3. MarkTechPost76

    Ricursive AI Releases Ripple-65B Preview: A 65B Dense Open-Weight Model Scoring 80.5 on Terminal-Bench 2.1

    Ricursive Intelligence released Ripple-65B Preview, a 65B dense open-weight model pretrained from scratch on 12T tokens with up to 1M context, shipping three Apache 2.0 checkpoints (instruction-tuned BF16, NVFP4, and BF16 base). Ricursive reports 80.5 on Terminal-Bench 2.1, the top self-reported score among US open-weight models, while trailing larger MoE rivals on SciCode (39.4) and DeepSWE 1.1 (44.3), with full R&D cost stated at $32M. Weights are on Hugging Face, with Together AI, GMI Cloud and OpenRouter listed at launch and an experimental NVFP4 build targeting NVIDIA DGX Spark.

    Why it matters: Useful for readers tracking open-weight agentic coding models, since it pairs self-reported benchmark claims with a detailed architecture, cost and deployment breakdown.

  4. Hugging Face Blog60

    Dharma Field Extractor: Any Document In, Structured Fields in JSON Out

    Dharma-AI introduces the Dharma Field Extractor, a 2-billion-parameter model built on DharmaOCR-LITE v1.1 with a LoRA adapter trained on 43,384 examples, which takes a document image plus either an open request or an exact schema and returns typed JSON, including null for fields not present. On a 546-document, 1,638-case benchmark it scored 0.834 field-level F1, above every frontier configuration tested (Claude Opus 5 at 0.816 to GPT-5.6 Luna at 0.741), while running self-hosted on a single L40S at a cost the company reports as 6 to 147 times lower than the APIs compared. The company frames the result as evidence that a small model trained close to a bounded task can beat a general frontier model on that task.

    Why it matters: Useful for teams weighing self-hosted specialized extraction against frontier APIs, though the benchmark and cost figures come from the vendor itself.