https://llmgram.app/live/hermes/ The releases, field reports and technical decisions shaping Nous Research and Hermes Agent. en Tue, 18 Aug 2026 22:15:44 +0000 LLMgram Live 1.0.0 / scs-add728fe68d910430b5d1bd0 https://llmgram.app/signal-core/releases/scs-add728fe68d910430b5d1bd0/live-signal-bundle.json https://x.com/Teknium/status/2089784829961458062 evt-788ebb50bb86cdd1dc3d Tue, 18 Aug 2026 18:41:44 +0000 Teknium announced that group chat bots can now have custom profile pictures and editable names, formalizing the concept of persistent agent profiles. This feature allows users to build a specialist bot once and reuse it indefinitely, transforming agents from stateless interactions into identifiable entities. model https://x.com/Teknium/status/2089657576300376099 evt-77b36a8d94674f25ef97 Tue, 18 Aug 2026 10:16:04 +0000 Nous Research has launched Hermes Agent, a free, MIT-licensed open-source agent harness developed with over 2,500 contributors. The project emphasizes complete optionality and capability, allowing developers to build custom agents without vendor lock-in. This release signals a strategic push to establish a standard open-source framework for agent orchestration. model https://www.marktechpost.com/2026/08/17/nous-research-hermes-bot-mode evt-c958d93ebd6eccc7e029 Tue, 18 Aug 2026 02:05:28 +0000 Nous Research has introduced Bot Mode for its open-source Hermes Agent, replacing the traditional session list with a roster of named bots. Each bot functions as a distinct profile with isolated memory, skills, and model configurations, and is now enabled by default in Hermes Desktop. This change allows users to maintain multiple specialized agents simultaneously rather than switching between generic sessions. model https://x.com/Teknium/status/2089499071547588922 evt-2aa08c5a20d4682fb76a Mon, 17 Aug 2026 23:46:14 +0000 Teknium confirmed that a dedicated mobile app for Hermes is in development but lacks a fixed release date. This move is intended to replace reliance on Telegram or Discord as the primary orchestration layer, allowing users to run and manage agents natively within the Hermes ecosystem. product https://x.com/Teknium/status/2089430781668303090 evt-95440c21aeb9dfdf5e91 Mon, 17 Aug 2026 19:14:52 +0000 Teknium has reintroduced Bot Mode for the Hermes Agent, a feature that allows users to maintain persistent chat profiles ('bots') with specific jobs and descriptions. Unlike standard sessions, these bots can communicate with one another, suggesting a move toward decentralized or networked agent architectures within a single desktop environment. research https://x.com/Teknium/status/2089267281847304522 evt-3fc9a75920a231b2c68d Mon, 17 Aug 2026 08:25:11 +0000 Teknium reports that GitHub has experienced approximately ten outages in a few days, preventing the merging of pull requests. This instability disrupts the continuous integration pipelines essential for modern AI-assisted software development and rapid iteration cycles. other https://x.com/NousResearch/status/2089186329431265599 evt-081b8cfd0c5ff1def30c Mon, 17 Aug 2026 03:03:30 +0000 Nous Research publicly acknowledged the developers and contributors behind the Hermes Agent, highlighting the collaborative effort required to maintain a world-class open-source tool. This move underscores the critical role of community engagement in the longevity and quality of AI agent frameworks. other https://x.com/Teknium/status/2089186056973562102 evt-3a0881e582a23ff844f4 Mon, 17 Aug 2026 03:02:25 +0000 Teknium announced that the Hermes Agent project has reached 2,500 contributors, highlighting a massive surge in community participation. This milestone indicates that the project is moving beyond initial development into a phase of heavy integration and review, where the volume of external contributions is outpacing internal capacity. other https://x.com/Teknium/status/2089163355261214830 evt-29615ee2b41b50ab1eb0 Mon, 17 Aug 2026 01:32:13 +0000 Teknium announced that the Hermes Agent platform has increased its user limit for ChatGPT and Codex subscriptions to 900,000. This move highlights the growing reliance on consumer-grade LLM subscriptions as the backend for complex agentic workflows, rather than relying solely on enterprise API keys. model https://x.com/Teknium/status/2089149835043807476 evt-0b19d733b6cfecdb5c49 Mon, 17 Aug 2026 00:38:29 +0000 Teknium highlights the ability to install the Hermes agent and run 'bots mode' inside Grok, suggesting a new layer of agentic orchestration. This indicates that major LLM providers are beginning to support or tolerate external agent frameworks that extend their core capabilities beyond simple text generation. model https://x.com/Teknium/status/2089144450266382828 evt-e6e42e78ab320ef13083 Mon, 17 Aug 2026 00:17:06 +0000 Teknium announced that he will update the Hermes agent to support a specific behavior following a recent error. This indicates that the Hermes framework is being adapted to handle edge cases or changes in model outputs that previously caused failures. model https://x.com/Teknium/status/2089121005168189539 evt-95b5c9a0de414c2a9ab2 Sun, 16 Aug 2026 22:43:56 +0000 Teknium clarified that the server-side maximum for Codex subscriptions is approximately 360K tokens, not 1M. In response, the Hermes Agent has raised its context window to 350K for Codex users to maximize current utility. The team plans to expand this to 1M tokens immediately if OpenAI exposes that capability. model https://x.com/Teknium/status/2089098159217918249 evt-686f81bed44a1a919c7f Sun, 16 Aug 2026 21:13:09 +0000 Teknium announced that Bots Mode is being packaged for the Hermes Agent Desktop release, highlighting a rapid development pace. This move integrates autonomous agent capabilities into a desktop environment, aiming to provide users with more sophisticated, hands-free AI interactions. model https://x.com/Teknium/status/2089036818624979191 evt-5ce0f0e26c19888310a5 Sun, 16 Aug 2026 17:09:24 +0000 Teknium announced that session loading times in Hermes Agent Desktop have been significantly reduced. This optimization targets the initial state hydration process, which is often a major friction point in complex agent interfaces. other https://x.com/Teknium/status/2088912079558439041 evt-8e7a08bbaf789da90d73 Sun, 16 Aug 2026 08:53:44 +0000 Teknium reports that Qwen 3.8:27b has become the primary orchestrator in the Hermes Agent, running effectively on a 24GB M4 Mini machine. This demonstrates that high-density intelligence is no longer restricted to cloud-based APIs or high-end data center GPUs, enabling robust local execution of complex multi-step tasks. model https://x.com/Teknium/status/2088874095345856541 evt-ea60ecdbd69cea77ef24 Sun, 16 Aug 2026 06:22:48 +0000 Teknium announced a significant update to the Hermes Agent Desktop, introducing the ability to scope skills, tools, and MCPs to specific profiles or bots. Additionally, users can now install skills directly via a built-in browser, streamlining the extension process. product https://x.com/Teknium/status/2088706343746310201 evt-21cae27b8ad4d2223eb6 Sat, 15 Aug 2026 19:16:13 +0000 Teknium announced that Bot Mode, previously accessible only via a quoted tweet, will be integrated into Hermes Desktop for all users by Monday. This move democratizes access to their agentic capabilities, moving beyond early-adopter testing to a stable product release. research https://news.google.com/rss/articles/CBMiaEFVX3lxTE1BNFppc0J4ZGRWZE5FamhEdkhvR0luVDh0cnFzRWN0WGN1SzVycG9iT2I1Qm9VTWxXSlNzVm1iU2lORXQzeWxkLVhLb00wd0dSamdxYWxkOUtaTXUwWF9zVEdjTXVadHA3?oc=5 evt-ed672fa604f46749bd91 Sat, 15 Aug 2026 07:23:49 +0000 Nous Research has opened a position for a Machine Learning Engineer focused on evaluations. This indicates a maturing approach to model development where systematic testing and benchmarking are prioritized alongside raw capability scaling. research https://x.com/Teknium/status/2088488103380205907 evt-1890aa45d1830e3628ea Sat, 15 Aug 2026 04:49:00 +0000 The post confirms active development and preparation for a specific feature release ('Bots Mode') in the Hermes Agent GUI, with slight uncertainty regarding the exact timing. model https://x.com/Teknium/status/2088471683762119117 evt-c46ccfe15586beb82f7d Sat, 15 Aug 2026 03:43:45 +0000 Teknium reported a scam email campaign impersonating GitHub, urging users not to click links. The post highlights an active threat vector where attackers exploit trust in major developer platforms to compromise infrastructure. model https://x.com/NousResearch/status/2088395070059770061 evt-e0855196a31f3c48c7fa Fri, 14 Aug 2026 22:39:19 +0000 Nous Research announced that Hermes Desktop can now connect to a Hermes Cloud agent via a settings gateway. This allows users to deploy agents that continue processing tasks in the cloud even after the local laptop is closed, providing remote accessibility and persistence. research https://x.com/Teknium/status/2088368313974047165 evt-57e11d320058dc75c318 Fri, 14 Aug 2026 20:53:00 +0000 Teknium announces the addition of a '/loop' command to the Hermes tool, enabling automated repeated actions at user-defined intervals. research https://x.com/NousResearch/status/2088367838977237029 evt-69d8a5d34f4f61fe0d00 Fri, 14 Aug 2026 20:51:07 +0000 Official announcement from Nous Research detailing a new '/loop' command for their Hermes tool that enables scheduled prompt execution. other https://x.com/Teknium/status/2088367346016776624 evt-c321c1c82633ae17fff2 Fri, 14 Aug 2026 20:49:09 +0000 Teknium reports that GitHub's unreliability, including site outages and CI failures, has forced OpenAI to develop internal infrastructure solutions. This highlights a critical dependency risk for AI companies that rely on third-party SaaS for core development workflows. other https://x.com/Teknium/status/2088328581953032238 evt-996b3abb5e7634df29d0 Fri, 14 Aug 2026 18:15:07 +0000 Teknium announced that the Hermes Agents interface now allows users to browse and attach specific skills from a central hub during the initial bot creation process. This feature streamlines the development workflow by decoupling skill selection from post-creation configuration, allowing for immediate functional definition of new agents. other https://x.com/NousResearch/status/2088319222417903864 evt-d71323e0c4d5d2fcca52 Fri, 14 Aug 2026 17:37:56 +0000 Nous Research announced the release of 'HERMES-4000,' a limited edition of 30 silkscreen prints featuring 23-karat red gold on black cotton rag. This physical artifact is tied to the HERMES model family, serving as a tangible brand ambassador for the AI lab's flagship product line. other https://x.com/Teknium/status/2088095297486610819 evt-d9cdbdf4ebc7d1a837e1 Fri, 14 Aug 2026 02:48:08 +0000 Teknium, a known researcher, is announcing the immediate availability of a new model for public testing, indicating a significant product release signal. product https://x.com/Teknium/status/2088073654643130833 evt-1ce9b40819106bf0898b Fri, 14 Aug 2026 01:22:08 +0000 Teknium has introduced 'Bot Mode' to the Hermes Desktop App, enabling users to manage multiple agent profiles and facilitating direct communication between them via an Agent Inbox. This update signals a strategic pivot from simple chat interactions to more complex, multi-agent workflows within the Hermes ecosystem. product https://x.com/Teknium/status/2088003994904113614 evt-9af15e56d6ec501efee5 Thu, 13 Aug 2026 20:45:20 +0000 Teknium, a key figure in the open-source AI community, is announcing a specific feature update for the Hermes Agent, indicating a product development signal. product https://x.com/Teknium/status/2087986084592709814 evt-e1750313f3c780f59420 Thu, 13 Aug 2026 19:34:10 +0000 Teknium announced a significant update to the Hermes agent framework, granting parent agents full lifecycle control over asynchronous sub-agents. This feature set allows for dynamic orchestration by reading live outputs, ending tasks early, and steering execution paths without waiting for full completion. other https://x.com/Teknium/status/2087972216566899100 evt-622fd458d8432b3c5a45 Thu, 13 Aug 2026 18:39:03 +0000 Teknium highlighted the addition of more free models to the Nous Portal, praising the team's output. This indicates a continuous stream of new open-weight releases rather than isolated events, reinforcing Nous's role as a key player in the open-source AI ecosystem. model https://x.com/NousResearch/status/2087970618167706023 evt-3a787c7ef0d33c1fc42c Thu, 13 Aug 2026 18:32:42 +0000 Nous Research is offering free access to Meituan's LongCat-2.0, a 1.6T-parameter Mixture-of-Experts (MoE) model, for one week. The model features a 1M token context window and scores 70.8 on Terminal-Bench 2.1, specifically optimized for ingesting entire codebases in a single pass. research https://x.com/Teknium/status/2087947369229009119 evt-91e9fee99cc38443b231 Thu, 13 Aug 2026 17:00:19 +0000 Teknium announced a massive expansion of the plugin capabilities for the Hermes agent framework, directing users to a specific GitHub issue for the full list of additions. This signals a strategic shift towards a more modular and extensible agent architecture within the NousResearch ecosystem. The update suggests a move from a monolithic agent design to a composable one, allowing for greater flexibility in tool integration. research https://x.com/Teknium/status/2087939423308746766 evt-6b470cb9fe8f32172771 Thu, 13 Aug 2026 16:28:45 +0000 The post highlights a significant market shift where improved model efficiency (DeepSeek Flash) is lowering barriers to entry for AI agent deployment, signaling a trend toward democratization. product https://x.com/Teknium/status/2087817045702582545 evt-ac841c59ec36553a42b5 Thu, 13 Aug 2026 08:22:28 +0000 Teknium highlights a showcase by @tonbistudio of Hermes Agent's Desktop plugins, emphasizing their positioning, form factors, and limitations. The demonstration includes a practical walkthrough of plugin creation and a casual gaming segment, indicating a focus on developer experience and usability. product https://x.com/imbabybrooklyn/status/2087737281683620196 evt-739babbebb2f1756b2fb Thu, 13 Aug 2026 03:05:30 +0000 Exact author post demonstrates an Inbox-style Hermes Agent Desktop feature. product https://x.com/Teknium/status/2087686461822996905 evt-d20216ab2b8ecbde39d5 Wed, 12 Aug 2026 23:43:34 +0000 Teknium released a new optional skill for the Hermes agent that monitors API calls made during web operations and automatically generates a static API client. This allows agents or scripts to interact with the site via the generated client in the future, bypassing the need for repeated web scraping or navigation. product https://x.com/NousResearch/status/2087592096769147377 evt-21128d7c0b2c0fabe151 Wed, 12 Aug 2026 17:28:36 +0000 Nous Research announced that Hermes profiles can now be saved as single files containing skills, memory, persona, crons, plugins, settings, and desktop themes, with credentials stripped for security. Users can then /import this file to instantly restore the complete profile configuration on a new instance. product https://x.com/Teknium/status/2087253636477080038 evt-3051e82cebfba2de4072 Tue, 11 Aug 2026 19:03:40 +0000 Tommy Tibo, a prominent figure in the open-weight community, announced that the 20% discount on all Nous Research models is extended for two more weeks. This promotion is tied to their token-based access portal (portal.nousresearch.com), encouraging users to acquire cheap compute tokens. The extension suggests the company is maintaining momentum in user acquisition and usage volume post-launch. model https://x.com/NousResearch/status/2087250500110450863 evt-f0c3f6b4f74d301b4b9f Tue, 11 Aug 2026 18:51:13 +0000 Nous Research announced an extension of its 20% discount on all models, including expensive frontier options, for an additional two weeks via its Nous Portal. This pricing strategy aims to lower the barrier to entry for high-performance models, encouraging broader adoption and increased token usage among developers and researchers. model https://x.com/NousResearch/status/2087197502415974634 evt-ac5f752598c22b2c1153 Tue, 11 Aug 2026 15:20:37 +0000 Upstage has released Solar Pro 4, an agentic model optimized for multi-step tasks across long documents, now available for free on the Nous Portal for one week. The model distinguishes itself by explicitly recognizing when it lacks sufficient information instead of inventing answers, a critical trait for production agents. It is priced at $0.30/$1.20 per million tokens, with a significant discount for cached tokens. model https://x.com/NousResearch/status/2086988904150798541 evt-6fab64368a870876ee62 Tue, 11 Aug 2026 01:31:43 +0000 Nous Research concluded its FLUX 3 short film contest, highlighting the public preview of the FLUX 3 model available via Hermes Agent and their paid Portal subscription. The contest winners received significant value (credits, hardware, subscriptions), signaling a strategy to drive paid adoption through engaging, creative use cases rather than just raw technical benchmarks. research https://x.com/Teknium/status/2086975696786829471 evt-e3cf7cf9cc9910c286e4 Tue, 11 Aug 2026 00:39:14 +0000 Teknium announced the live availability of the 'Hermes Pixel Office' VSCode extension and GitHub plugin, enabling developers to visualize Hermes agents working in real-time. This release marks a strategic pivot for Hermes, moving beyond just providing LLM weights to offering developer tooling that enhances transparency and debugging for agent-based workflows. product https://x.com/Teknium/status/2086961142329725233 evt-2602aee071cf35816b35 Mon, 10 Aug 2026 23:41:24 +0000 Teknium posted a cryptic update claiming success after 'one session' with a promise to share setup links, but the attached content is merely a client-side script for displaying dates and times in a specific timezone. This suggests the core value proposition (the model or agent) is hidden behind a link or visual element not captured in the text excerpt. model https://x.com/Teknium/status/2086882821910782270 evt-7647dbf0aaf172db2162 Mon, 10 Aug 2026 18:30:11 +0000 The post shares a concrete technical optimization strategy for agent tool usage, indicating a practical research or engineering finding. research https://x.com/Teknium/status/2086881909209252209 evt-9ebd0e0db6f1f638db84 Mon, 10 Aug 2026 18:26:34 +0000 Teknium highlights that the Browser-Use CLI 3.0 update has successfully reduced token consumption by 60% for browser-based AI tasks. This improvement is attributed to a redesigned backend that optimizes how the browser state is processed and sent to the LLM. The result is a more economical approach to running autonomous agents that interact with web interfaces. research https://x.com/NousResearch/status/2086881660658663469 evt-1802bc6fb8355982d952 Mon, 10 Aug 2026 18:25:34 +0000 Hermes introduces a 'Browser Use' mode that replaces multiple specialized browser tools with a single agent capable of writing and executing scripts. This shift from a schema-heavy, click-by-click interaction model to a script-based approach significantly reduces token consumption, as demonstrated by a 48% reduction in test environments. The move prioritizes efficiency and scalability in web automation tasks. product https://x.com/Teknium/status/2086712331720183842 evt-35ecfe5934c8c14bd6e7 Mon, 10 Aug 2026 07:12:43 +0000 Teknium, a prominent open-weight model developer, announced 'Hermes Agent' on August 10, 2026, positioning it as a tool to 'make anything real,' implying a shift from passive LLM usage to active, result-oriented agents. The timestamp suggests this is a recent development in the post-2025 landscape where agentic capabilities became the primary differentiator for open-source models. The brevity of the post highlights a product launch or significant capability update rather than a theoretical paper. other https://x.com/Teknium/status/2086707217571750323 evt-3ee8318c4226018ab7a1 Mon, 10 Aug 2026 06:52:24 +0000 Teknium announced that the Nous Portal now supports a significantly expanded list of image and video generation models, moving beyond its previous limitations. This update allows users to access a wider variety of generative AI capabilities through a single interface, reducing the friction of managing multiple model endpoints. The move positions Nous Portal as a more versatile hub for developers and researchers experimenting with different generative architectures. model https://x.com/Teknium/status/2086702328024125926 evt-1585cb57c3646fa99120 Mon, 10 Aug 2026 06:32:58 +0000 Teknium has released improvements to the Hermes Agent, specifically a rebuilt 'read tool' designed to save users tokens and time. This component was engineered from scratch for the v1 release to optimize performance for open models, addressing limitations found in previous versions. model https://x.com/NousResearch/status/2086621731851026912 evt-0e72e311469b0f973d25 Mon, 10 Aug 2026 01:12:43 +0000 Nous Research announced a three-day extension of a 90% discount on the DeepSeek V4 Flash model, directing users to their portal for access. This move capitalizes on the 'popular demand' for efficient, high-performance open-weight models, aiming to rapidly expand the user base and ecosystem integration for their specific build of the V4 architecture. research https://x.com/Teknium/status/2086568600937357696 evt-3c8e3155dd9e533bfe1b Sun, 09 Aug 2026 21:41:35 +0000 Teknium announced a new optional skill for the Hermes AI ecosystem that enables fast, multi-language codebase searching using AST-grep. The feature is distributed as an installable module, allowing users to extend the base model's capabilities on demand rather than relying on a monolithic update. other https://vercel.com/changelog/vercel-ai-gateway-and-vercel-sandbox-now-available-on-hermes-agent evt-80c665702fd4deedbcdb Fri, 07 Aug 2026 19:00:00 +0000 Hermes Agent now supports Vercel AI Gateway as its inference layer, providing access to 200+ models with zero token markup and centralized dashboard visibility. Simultaneously, it leverages Vercel Sandbox microVMs to execute agent commands in isolated environments, enhancing security and reproducibility. model https://x.com/NousResearch/status/2085184069302935999 evt-8ab3f6028e6b181db7a2 Thu, 06 Aug 2026 01:59:57 +0000 Nous Research announced 'Actual', a local inference stack designed to run efficiently on personal compute with minimal CPU impact. The tool emphasizes multi-platform portability and is marketed as a complementary solution for users running Hermes Agents locally. This release targets the growing demand for efficient, on-device AI execution outside of heavy GPU-dependent cloud setups. research https://x.com/Alibaba_Qwen/status/2084683919937634507 evt-f3e3006237b8899eda94 Tue, 04 Aug 2026 16:52:32 +0000 Qwen has announced the availability of its Qwen3.8-Max model within the Hermes Agent ecosystem, signaling a push to demonstrate its model's efficacy in autonomous workflows. This release allows developers to immediately test whether the model's advanced reasoning holds up when wrapped in rigid tool-calling patterns typical of agent architectures. model https://pub.towardsai.net/openclaw-vs-hermes-agent-the-honest-comparison-nobodys-given-you-yet-6c3ef10d518b?source=rss----98111c9905da---4 evt-015a447aa140811c34b8 Sun, 02 Aug 2026 23:01:02 +0000 Peter Steinberger’s OpenClaw, which rapidly became GitHub’s most-starred repository, is compared against Hermes Agent with a focus on security timelines and production readiness rather than just feature sets. The piece argues that existing comparisons overlook critical details regarding credential handling and real-world migration paths, which are vital for teams moving beyond experimental stages. model https://pub.towardsai.net/becoming-a-top-1-hermes-agent-user-the-complete-playbook-no-one-else-is-sharing-901c7757ed1c?source=rss----98111c9905da---4 evt-c44260b744dfe6fde460 Sun, 02 Aug 2026 22:01:01 +0000 Nous Research's Hermes Agent, released in February 2026, introduces a framework where the AI extracts skills by observing user behavior rather than relying solely on predefined configurations. This article details the operational playbook for running Hermes as persistent infrastructure, highlighting its ability to autonomously generate capabilities from repetitive tasks. other https://www.marktechpost.com/2026/07/31/nous-research-ships-three-integration-paths-for-hermes-agent-and-buzz-blocks-open-source-nostr-workspace-for-humans-and-agents evt-12fb5620f9402c272f5d Fri, 31 Jul 2026 09:08:13 +0000 Nous Research has enabled Hermes Agent support for Buzz, Block's open-source Nostr workspace, offering three integration paths: Desktop runtime, relay bridge, and a native gateway. This allows AI agents to operate within Nostr channels while preserving specific model attributes like memory, skills, and cron delivery. model https://github.com/uzairansaruzi/hermex evt-e9f0e35d7e3c49731b35 Thu, 16 Jul 2026 08:37:09 +0000 Hermes now has an iPhone client: https://github.com/uzairansaruzi/hermex Enthusiasts have built Hermex, a native iPhone app using SwiftUI that allows you to fully manage your self-hosted Hermes Agent directly from your phone. 🚬 product https://www.reddit.com/r/LocalLLaMA/comments/1uvypn9/hermes_agent_with_local_llms evt-23b7a8473f765cf60792 Tue, 14 Jul 2026 04:44:18 +0000 A user with a high-end consumer GPU setup (4090 + 5060ti) is evaluating the financial viability of switching from a $20/day hosted agent service to a local open-source model like Qwen 2.5-32B or Hermes. The core challenge identified is balancing inference speed (TPS) and loop stability against the hardware limits of a dual-GPU consumer rig. model https://www.techmeme.com/260713/p33 evt-3eb7633ac9d0d7a8ec56 Mon, 13 Jul 2026 23:45:02 +0000 Nous Research, creator of the Hermes open-source agent, is raising over $75M led by Robot Ventures at a $1.5B valuation. This funding round validates the commercial viability of specialized open-source agents competing directly with closed-source rivals like OpenClaw. research https://techcrunch.com/2026/07/13/hermes-agent-maker-nous-research-in-talks-for-new-funding-at-1-5b-valuation evt-2adbe67cb2699f3b1a0b Mon, 13 Jul 2026 23:31:18 +0000 Nous Research is raising at least $75 million in new funding led by Robot, with participation from USV, valuing the company at $1.5 billion. This capital injection supports their expansion beyond open-weight models into the Hermes agent ecosystem, positioning them as a key player in the agentic AI layer. research https://x.com/NVIDIAAI/status/2074207653367058528 evt-62a6994ac6c9b4f2893d Mon, 06 Jul 2026 19:03:36 +0000 Chess legend Vidit Gujrati participated in the Hermes Agent Accelerated Business Hackathon, utilizing the Nemotron model to construct an AI agent. The submission highlights the integration of high-level domain expertise with accessible AI infrastructure provided by NousResearch and Stripe. research