Ollama Blog5858Ollama's Pro, Max, and Team plans now adopt a transparent token-based pricing, including usage limits per month. The Pro plan is $20 per month (including a $60 credit), Max is $100 per month (including a $300 credit), and Team is $500 per month (including a $1000 shared credit). The free plan offers a small initial model credit, which can be expanded on a pay-as-you-go basis.
Dify BlogDify launches a new intelligent agent (New Agent), where each agent has independent configuration and lifecycle, supporting one-time construction and reuse across workflows. This agent can be published as a web application or exposed via API, featuring three core functions: the ability to build and optimize with real tasks, unified cross-workflow management to reduce repetitive maintenance, and full observability for single-run tracking and long-term monitoring.
OpenHands Blog6565OpenHands Enterprise users can now directly trigger AI development workflows from Jira Cloud projects. Mention @openhands or add the openhands tag in a Jira issue to start the workflow.
SGLang releasesSGLang v0.5.18 released, adding multimodal models such as Muse and Glimmer, and models like LingBot-Video-MoE and LongCat-Image. Qwen3-32B launch speed improved 2.38 times to 35.6 seconds on H100. TP LMHead bandwidth optimization reduced DeepSeek-V4-Pro B200 decoding speed from 320 microseconds to 169 microseconds.
Qoder Blog6565Qoder has obtained verifiable assurance through independent SOC 2 Type II audit, supporting enterprise AI coding applications; the enterprise version is priced at $3/M tokens, with an 8k context window, and supports Agent SDK and Cloud Agents.
Qoder Blog6565The child aims to reduce the end-to-end requirements for a week to a single day delivery through Qoder, building an AI platform Zhishu that runs 400 scenarios and 300 agents.
Microsoft AI NewsMAI-Cyber-1-Flash is designed for security tasks, capable of efficiently handling 90% of tasks, allowing MDASH to use more expensive GPT-5.4 for 10% of complex tasks, reducing costs by 50%. Perception, as a new intelligent security system, is also released, will use MAI-Cyber-1-Flash to expand more security workflows.
Factory News7575Factory's Agent Effectiveness feature is now live, linking data with delivery cycles, work intentions, and delivered outcomes to help engineering leaders understand delivery acceleration points, robot work content, and how usage maps to actual output. This feature provides an operational view of AI investment return by connecting Droid usage with delivery results, addressing questions about delivery speed changes, profit concentration points, and whether new capabilities are used for priority work.
Prime Intellect BlogPrime Flash MoE is a set of CUDA kernels optimized for Blackwell, which significantly reduces memory traffic by avoiding the explicit creation of certain intermediate tensors and the explicit creation of activation tensors in fusion configurations.
LangChain BlogLangSmith Bring Your Own Cloud (BYOC) is now officially available on AWS, offering businesses a hosted LangSmith deployment that can run in their AWS accounts and VPCs.
Liquid AI BlogMacPaw and Liquid AI announce a strategic partnership to combine Liquid's base model with MacPaw's Elix and Mnemos technologies, creating the first Mac assistant Eney that integrates on-device intelligence, persistent memory, and native task execution, expected to launch later this year and expand to more products afterward
LangChain BlogFeatured6262LangSmith LLM Gateway has entered public beta, serving as a centralized governance layer between agents and models, uniformly executing spending limits, rate limiting, model fallback, and sensitive data deidentification. Spending and rate limiting can be set at four levels: organization, workspace, API key, and user. When the limit is hit, the agent receives a 402 response; fallback rules can take effect uniformly across models and hosting providers.
Why it matters: The original text listed the specific control levels of the gateway in terms of cost, rate limiting, fallback, and data masking, which can help determine at which level the governance of the production intelligent agent should be placed.
Microsoft AI NewsThe MAI-Cyber-1-Flash model achieved a 96% score in the CyberGym evaluation, with a cost of only 50% of the current MDASH best solution, optimized specifically for MDASH actuators.
Mastra Blog7575Mastra Factory has been released, a system that enables agents to handle from problem processing to software production, supporting code writing, validation, release, documentation updates, and monitoring, integrated with GitHub, Linear, and Slack, providing typed workflows, memory, scheduling, tools, and observability features.
Microsoft AI NewsThe Hill-climbing MAI model uses 10% fewer tokens in GitHub Copilot, has a 10% higher code acceptance rate than GPT 5.4 Mini, and an 11% higher code acceptance rate than Claude Haiku 4.5, with a 6% and 11% higher user retention rate, respectively.
Dify Blog7575Dify v1.15.0 releases command-line tool difyctl, enabling web applications to be directly invoked by automation scripts, CI, and AI Agents. Users can install Agent skills via `difyctl skills install`, allowing Codex, Claude Code, and other intelligent agents to automatically discover applications, identify parameters, invoke Workflows, and handle pause/resume.
Qoder Blog6565Qoder Knowledge Engine transforms engineering knowledge into reusable context, enhancing the accuracy, stability, and execution efficiency of AI agents.
Dify Blog6565Dify Cloud helps IT teams track token usage, message volume, and active users through unified monitoring and logging features, identify and optimize inefficient processes; its model-agnostic feature allows configuring OpenAI, Anthropic, Google, and open-source models for different nodes in a single workflow, matching models to tasks to reduce costs.
Qoder Blog3232Qoder launches Qoder Code Security, ensuring security from the first line of code; the product is integrated through Agent SDK and Cloud Agents, supporting multiple usage methods.
Dify BlogQubrid AI provides builders of Dify with unified API access to leading open-source and cutting-edge models such as DeepSeek, Kimi, Qwen, MiniMax, and GLM through a single plugin
Qoder Blog7575Alibaba Qoder New Model CaseOne Paragraph Summons 13 "Programmers" By Automatically Decomposing Requirements and Assigning Multiple Expert Agents to Develop in Parallel, You Can Build and Rebuild Personal Blogs With Just One Paragraph of Text.
Qoder Blog3030Qoder's expert mode, through 30 minutes of parallel work, helps users quickly assemble a multi-agent expert team and deliver merge main branch functionality. This mode supports creating AI employees and can be integrated via Agent SDK and cloud proxy.
Qoder Blog6565Hisense has built a complete AI native development system with Qoder, with new system development efficiency improved 3-4 times.
Microsoft AI NewsMAI's audio model Ode Poetry recommends and plays real human reading recordings of poems for users worldwide, providing a digital poetry experience. The system uses MAI-Transcribe to understand user language in real-time, achieving the lowest word error rate of 4.86% in the FLEURS benchmark test; MAI-Voice can faithfully reproduce the unique voice of William Sieghart. The system combines William's literary analysis experience, recommends poems through custom logic and tool calls, and iteratively optimizes for a safe, high-quality user experience.
Dify BlogDify has released its first official command-line tool, difyctl, which supports calling Dify applications and workflows in terminals, scripts, CI pipelines, or intelligent agents like Claude Code, Codex, and Cursor to achieve automation and integration. difyctl is cross-version across Enterprise, Cloud, and Community editions, providing a unified command set without the need for custom integration code. It supports login via Dify accounts or external SSO and is available in the Enterprise version.
Cognition BlogDevin Fusion maintains cutting-edge and Fable 5-level performance at 60% lower cost on the new coding benchmark FrontierCode, primarily through a dual-agents architecture that runs cutting-edge models in parallel with cost-effective 'assistant' models. This architecture avoids benchmark overfitting and enhances task generality through dynamic in-session routing and deeply tuned interaction patterns, showing the most significant efficiency improvement in implementation-oriented sessions. Users can try the preview version via app.devin.ai/signup.
Qoder Blog1717Eccang breaks the boundary between technical and non-technical teams using Qoder, transforming employees from 'tool users' to 'agent managers'. Qoder supports cross-team collaboration, helping companies achieve intelligent upgrades in R&D and operations.
Dify BlogDify has released MongoDB Atlas and Voyage AI plugins on the Dify Marketplace through collaboration with MongoDB and Voyage AI, enabling Dify builders to directly connect to MongoDB data, use Atlas Vector Search for semantic search, and improve result quality through Voyage AI embedding and reordering.
Qoder Blog5555Qoder Terminal Sandbox is released, providing a cross-platform native command isolation execution environment for AI agents on Windows, macOS, and Linux, supporting bounded agents.
Factory NewsDroid runs STRIDE-based security reviews on every non-draft PR, attaching severity level, CWE reference number, explanation, and repair suggestions to discovered vulnerabilities, and publishing them as inline comments directly on the code diff. This feature has been running in production code repositories and has discovered vulnerabilities such as CVE-2026-42876. Droid filters false positives by building lightweight threat models, scanning for vulnerabilities, and verifying candidate discoveries, generating only one summary comment per PR.
Prime Intellect BlogPrime Intellect Lab fully supports fine-tuning on Nemotron 3 Ultra (550B parameter MoE model), providing managed RL training for Hikvision B200/B300, 2500+ development environments, and SWE-bench Verified benchmark tests.
ByteDance Seed · Hugging Face new models2525ByteDance Seed/TaskMem model release, with 31B parameter scale, using BF16 tensor type, but no inference providers are deployed.
Prime Intellect Blog7575Prime Intellect has released Hosted Evaluations, providing scalable infrastructure for users to test models in self-evaluation environments. The service supports running evaluations on the hundreds of community environments already available in Environments Hub, with users able to start them via the platform, CLI, or .toml files. It supports benchmark tests like GPQA, SWE-bench, and results can be published to corresponding leaderboards with transparent display. This aims to address the infrastructure needs of modern evaluations and improve testing efficiency and reproducibility.
Cognition BlogThe Cognition team is promoting asynchronous software engineering, with more Devin triggered by events and automation, requiring end-to-end cloud validation to ensure results are reliable. Devin Review can automatically fix code review issues, but needs further testing and validation. Devin runs in cloud virtual machines, controlled with a mouse and keyboard, able to test its own work, and can run 10-20 instances in parallel, saving local validation time. It improves reliability through test plans, deterministic scripts, and skill libraries.
Dify Blog7575Dify is an LLM application development platform that enables marketers to build structured, repeatable workflows containing LLMs, vector databases, external APIs, and logical steps through visual drag-and-drop. It supports multi-model generation (such as GPT-4, Claude, Gemini) for parallel content creation, automatic scoring and lead routing, knowledge base for marketing with Q&A, differing from tools like ChatGPT and Claude that only support single-model conversations.
Qoder Blog6565Qoder provides a professional AI workspace, supporting tasks beyond coding, not just a chat box; integrates Agent SDK and cloud proxy, context length 8k, parameter scale 175B, speed increased 3.5x, price $3/ million tokens.
