Nous Research 發布 Hermes Desktop:Hermes Agent v0.15.2 的原生跨平臺前端,支援串流工具輸出
重點摘要
Nous Research 推出 Hermes Desktop 公開預覽版,這是一款原生支援 macOS、Windows 與 Linux 的應用程式,為開源 Hermes Agent 提供圖形化介面。在此之前,使用者只能透過命令列介面(CLI)與訊息閘道來執行 Hermes。目前版本為 Hermes Agent v0.15.2。根據 Nous Research 的文件,桌面版沿用相同的代理核心,並與 CLI 及閘道共享配置、API 金鑰、會話、技能與記憶體。桌面版僅是同一代理的另一個操作介面,並非分支版本。Hermes Desktop 是什麼?Hermes Agent 是一個自主式 AI 代理,並非綁定於編輯器的程式碼助手。它能執行任務、呼叫工具,並跨會話保持狀態。所謂「代理」,指的是在循環中進行規劃、行動與觀察的模型。Hermes Desktop 便是建構於該代理之上的圖形化前端。
Nous Research has released Hermes Desktop in public preview. It is a native application for macOS, Windows, and Linux. It gives the open-source Hermes Agent a graphical interface. Until now, users ran Hermes through a CLI and messaging gateways. The current build is Hermes Agent v0.15.2. Per Nous Research’s documentation, the desktop reuses the same agent core. It shares configuration, API keys, sessions, skills, and memory with the CLI and gateway. The desktop is another surface over one agent, not a fork. What is Hermes Desktop Hermes Agent is an autonomous AI agent. It is not a coding copilot tied to an editor. It runs tasks, calls tools, and keeps state across sessions. An agent here means a model that plans, acts, and observes in a loop. Hermes Desktop is a GUI on top of that same agent core. It needs no terminal to use. The window shows streaming responses and live tool activity. A right-hand pane previews web pages, files, and tool outputs. It also includes a file browser, voice input and output, and a settings UI. Sessions are shared across surfaces. A conversation started in the desktop resumes in the CLI or TUI. The reverse also works, because state is not duplicated. macOS and Windows offer direct installers. Linux installs from the terminal on any distribution. An install script with an --include-desktop flag builds the app against an existing install. The Closed Learning Loop Nous research team describes Hermes as having a closed learning loop. This is what separates it from a simple chat wrapper. After a complex task, the agent writes a reusable skill. Those skills then self-improve during later use. Memory is persistent and agent-curated, with periodic nudges to save knowledge. Cross-session recall uses FTS5 session search with LLM summarization. User modeling runs through Honcho dialectic user modeling. In practice, longer use means more retained context and reuse. Skills follow the agentskills.io open standard. How It Connects, Schedules, and Sandboxes Hermes runs across messaging platforms from one gateway. The desktop lists Telegram, Discord, Slack, WhatsApp, Signal, Email, and CLI. You can start a task on one platform and continue on another. Scheduling uses natural language for reports, backups, and briefings. These run unattended through the gateway on a built-in cron scheduler. Delegation spawns isolated subagents with their own conversations and terminals. A subagent is a separate worker that handles one job. Python RPC scripts collapse multi-step pipelines into zero-context-cost turns. Execution is sandboxed. The desktop lists five backends: local, Docker, SSH, Singularity, and Modal. It applies container hardening and namespace isolation. Namespace isolation limits what a running process can see or touch. Built-in tools include web search, browser automation, vision, image generation, text-to-speech, and multi-model reasoning. Hermes also connects external tools through MCP. MCP is the Model Context Protocol, a standard for tool integration. Nous Portal and the Tool Gateway Hermes works with any provider, so API keys are optional. Nous Portal bundles them under one subscription instead. Portal tiers are Free, Plus, Super, and Ultra. Paid tiers include monthly credits and access to 300+ models. They also include built-in tool use. The Tool Gateway routes several tools through one account. Web search uses Firecrawl and image generation uses FAL. Text-to-speech uses OpenAI and the cloud browser uses Browser Use. The next evolution of Hermes Agent is here! Introducing Hermes Desktop: everything you love about Hermes, now native on your machine.First demoed in Jensen's GTC keynote, it's now in public preview. pic.twitter.com/8ND1k8hyaz— Nous Research (@NousResearch) June 2, 2026 Strengths and Questions Strengths: Native installers remove the terminal requirement for most users Streaming output and previews make tool calls easier to inspect Persistent memory and self-improving skills reduce repeated instructions Model-agnostic design avoids lock-in to a single provider The MIT license allows audit, self-hosting, and modification Questions: The product is in public preview, so expect rough edges Autonomous memory and scheduling raise oversight and review questions The Linux desktop still installs through the terminal Broad capability means a steeper learning curve for beginners Key Takeaways Nous Research released Hermes Desktop in public preview, a native macOS, Windows, and Linux app for its open-source Hermes Agent. The GUI shares one agent core, configuration, API keys, sessions, skills, and memory with the CLI and gateway; sessions resume across surfaces. It runs no-terminal with streaming tool output, a side-by-side preview pane, file browser, voice I/O, and a settings UI. Hermes is model-agnostic and MIT-licensed, working with Nous Portal, OpenRouter, OpenAI, or any compatible endpoint. The current build is Hermes Agent v0.15.2, backed by a closed learning loop, MCP tool support, and five sandbox backends. Check out the Project here. Also, feel free to follow us on Twitter and don’t forget to join our 150k+ ML SubReddit and Subscribe to our Newsletter. Wait! are you on telegram? now you can join us on telegram as well. Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.? Connect with us The post Nous Research Releases Hermes Desktop: A Native Cross-Platform Front End for Hermes Agent v0.15.2 with Streaming Tool Output appeared first on MarkTechPost.
Related
相關文章
Liquid AI Introduces LFM2.5-Embedding-350M and LFM2.5-ColBERT-350M: Dense Bi-Encoder and Late-Interaction Models for Fast Multilingual Search Across 11 Languages
This week, Liquid AI released two new retrieval models. They are LFM2.5-ColBERT-350M and LFM2.5-Embedding-350M. Both hold 350M parameters. Both are the first bidirectional members of the LFM family. They build on LFM2.5-350M-Base, released in March. The pair targets fast multilingual and cross-lingual search across 11 languages. Their footprint is small enough to run almost anywhere. Both are available now on Hugging Face under the LFM Open License v1.0. LFM2.5 Retrievers The two models share one backbone but represent text differently. LFM2.5-Embedding-350M is a dense bi-encoder. It turns each document into a single vector. Pick it when you want the fastest search and the smallest, cheapest index. LFM2.5-ColBERT-350M is a late-interaction model. It converts each token into a vector rather
Perplexity Launches Brain, a Self-Improving Memory System That Builds a Context Graph of an Agent’s Work and Learns Overnight
Most AI memory remembers the user. It stores your preferences, your tastes, and your role. Perplexity is taking a different path. Today, Perplexity launched Brain, a self-improving memory system for its agent product, Computer. Brain does not focus on remembering you. It remembers what the agent did. That reframes what memory in AI is for. What is Perplexity‘s Brain Brain is a self-improving memory system. It builds a context graph of the work Computer performs. At set intervals, such as overnight, Brain reviews that graph. It then teaches itself how to do the work better. The idea is straightforward. The more work you do, the more efficient Brain makes your Computer. Brain is rolling out today to Perplexity Max and Enterprise Max subscribers in Research Preview. Two Axes of AI Memory Perp

智譜新高,MiniMax承壓,“大模型雙雄”命運殊途
這篇消息聚焦「智譜新高,MiniMax承壓,“大模型雙雄”命運殊途」。原始導語提到:大模型在被市場重新定價 從 AI 情報角度來看,這類內容值得關注其背後的技術進展、產品落地、產業競爭與後續市場影響。

華為昇騰 0 Day 支持智譜 GLM-5.2 模型,提供全面推理優化
華為昇騰 AI 宣佈在智譜開源 GLM-5.2 大模型當天即完成深度推理優化。通過 MOE 大融合算子、通信計算融合、高併發調度等七項關鍵技術,顯著提升編程和長程任務的處理效率,現已支持 A3 系列產品部署。#AI 大模型# #國產算力#
企業AI轉型再添利器:青雲科技算力雲接入 MiniMax-M3 模型
企業AI落地面臨高效低成本難題。青雲科技旗下基石智算平臺接入國產開源大模型MiniMax-M3,提供新算力支持。MiniMax-M3以卓越上下文處理能力等三大核心技術見長,依託自研架構,助企業便捷部署AI業務。
阿里開源統一科學大模型 LOGOS,僅用五十六分之一參數超越微軟
阿里 ATH-Token Foundry 聯閤中國人民大學高瓴人工智能學院開源科學基礎模型 LOGOS。該模型採用統一科學語法與純序列建模範式,在六大科學任務上匹配或超越傳統專用方法。其中 LOGOS-1B 僅 1B 參數,即展現出極高效率,性能超越參數量達 8×7B 的微軟模型。