Close Menu
    Facebook X (Twitter) Instagram
    Articles Stock
    • Home
    • Technology
    • AI
    • Pages
      • About ArticlesStock — AI & Technology Journalist
      • Contact us
      • Disclaimer For Articles Stock
      • Privacy Policy
      • Terms and Conditions
    Facebook X (Twitter) Instagram
    Articles Stock
    AI

    OpenClaw vs Hermes Agent: Why Nous Analysis’s Self-Enhancing Agent Now Leads OpenRouter’s International Rankings

    Naveed AhmadBy Naveed Ahmad10/05/2026Updated:10/05/2026No Comments9 Mins Read
    blog11 5


    The open-source AI agent area has a brand new chief. As of Could 10, 2026, Hermes Agent — constructed by Nous Analysis — has overtaken OpenClaw to carry the #1 place on OpenRouter’s global daily app and agent rankings. Hermes is at present producing 224 billion each day tokens on OpenRouter versus OpenClaw’s 186 billion, making it probably the most actively used open-source AI agent by present inference quantity.

    The milestone is extra important than a easy leaderboard swap. OpenClaw’s founder, Peter Steinberger, joined OpenAI in February 2026, whereas OpenClaw moved to an unbiased open-source basis with OpenAI as a sponsor.

    Two Completely different Bets on What an Agent Ought to Be

    The rivalry between Hermes and OpenClaw comes all the way down to a elementary architectural disagreement. OpenClaw is organized round a central WebSocket Gateway — a persistent routing layer that connects 50+ messaging channels (Telegram, Discord, Slack, WhatsApp, Sign, and extra) to an agent runtime. Its design optimizes for attain: what number of surfaces the agent can function throughout concurrently.

    Hermes Agent takes the alternative strategy. Constructed below an MIT license, it facilities on a “do, study, enhance” execution loop. After finishing a job, the agent enters a reflective part the place it analyzes its personal efficiency and autonomously generates reusable talent information for future use. Reminiscence is dealt with by way of three layers: a persistent snapshot of person and agent id, a SQLite FTS5 full-text search database of each previous session, and procedural talent information that seize repeatable job logic. The design is constructed for compounding worth over time — the longer you run Hermes, the extra optimized it turns into on your particular workflows.

    A Fast Launch Cadence

    Hermes has shipped a confirmed main launch repeatedly since its February 2026 launch. The v0.9.0 “Everywhere” release introduced Android/Termux assist, iMessage by way of BlueBubbles, WeChat and WeCom adapters, and an area net dashboard, pushing Hermes to 16 supported messaging platforms. The v0.11.0 “Interface” release delivered a full React/Ink TUI rewrite, native AWS Bedrock assist, 5 new inference paths together with NVIDIA NIM and Vercel ai-gateway, GPT-5.5 entry by way of Codex OAuth, and a seventeenth platform by way of QQBot — throughout 1,556 commits and 761 merged PRs.

    The v0.13.0 “Tenacity” release, shipped Could 7, 2026, is the present newest. It introduces Kanban as a sturdy multi-agent job board with heartbeat monitoring, zombie detection, and hallucination restoration; a /aim command that locks the agent on a goal throughout turns; Checkpoints v2 with actual state pruning; gateway auto-resume after restart; and Google Chat because the twentieth supported messaging platform.

    Safety: A Distinction Value Noting

    OpenClaw’s scale has include safety prices. CVE-2026-25253, assigned a CVSS rating of 8.8, uncovered the gateway to distant exploitation. In a four-day window in March 2026, 9 CVEs have been disclosed — one scoring 9.9. A Koi Safety audit of two,857 ClawHub expertise discovered 341 malicious entries, with 335 tied to a single marketing campaign, and broader third-party safety scans flagged over 800 suspicious entries throughout the identical interval. SecurityScorecard reported tens of 1000’s of publicly uncovered OpenClaw situations.

    Hermes Agent’s safety document is shorter by advantage of its age, however not clear. NVD lists a number of CVEs revealed April 27–29, 2026, together with CVE-2026-7113, a lacking authentication challenge within the webhooks endpoint of model 0.8.0, scoring 5.6 MEDIUM on CVSS 3.x and a couple of.9 LOW on CVSS 4.0. The v0.13.0 Tenacity launch addressed 8 P0 safety points, together with enabling redaction by default, guild-scoped Discord position allowlists, WhatsApp stranger rejection, and TOCTOU patches throughout auth.json and MCP OAuth flows.

    Migration and the Path Ahead

    For builders working OpenClaw who need to consider Hermes, the transition is designed to be low-friction. Hermes detects an present ~/.openclaw listing throughout setup and gives to import settings, recollections, expertise, and API keys mechanically. The hermes claw migrate command helps dry-run previews, selective migration presets, and battle overwrite controls.

    The 2 frameworks are additionally more and more being run in parallel, with OpenClaw dealing with orchestration and multi-channel routing whereas Hermes executes repeatable job loops — coordinated by way of the Agent Communication Protocol (ACP).

    The broader sign from the OpenRouter rankings is that the open-source agent market shouldn’t be consolidating round one software. It’s bifurcating round two totally different philosophies: breadth of attain versus depth of studying. Hermes Agent’s transfer to #1 on each day utilization means that, not less than for now, a significant portion of the developer group is betting on depth.

    224B
    Hermes Each day Tokens
    #1 on OpenRouter

    186B
    OpenClaw Each day Tokens
    #2 on OpenRouter

    6.35T
    Hermes All-Time
    vs 9.17T OpenClaw

    v0.13.0
    Newest Launch
    Could 7, 2026




    Head-to-Head

    Hermes Agent OpenClaw
    Creator Nous Analysis (MIT) Peter Steinberger / Basis (OpenAI sponsor)
    Launched February 2026 Late 2025
    GitHub Stars 114,000+ 370,000+
    Each day Tokens 224B (#1 OpenRouter) 186B (#2 OpenRouter)
    All-Time Tokens 6.35T 9.17T
    Structure Self-improving execution loop; do–study–enhance WebSocket Gateway; central routing controller
    Reminiscence SQLite FTS5 + persistent id + auto-generated expertise Markdown-based; human-authored expertise by way of ClawHub
    Talent Ecosystem 40+ built-in instruments; agentskills.io normal; auto-generates expertise 44,000+ ClawHub expertise (April 2026)
    Platforms 20 (incl. Google Chat, QQBot, WeChat, iMessage) 50+ channels
    Mannequin Help 200+ by way of OpenRouter; NVIDIA NIM; AWS Bedrock; Ollama; native OpenAI-compatible; OpenRouter; Ollama
    Newest Model v0.13.0 — Could 7, 2026 Energetic (LTS introduced Could 2026)
    CVE Depend 4 CVEs (Apr 27–29; have an effect on v0.8.0) 9 CVEs in 4 days (Mar 2026); one CVSS 9.9
    Migration hermes claw migrate imports OpenClaw config mechanically —

    Hermes Agent Launch Cadence

    v0.13.0 — The Tenacity Launch
    Newest
    Could 7, 2026

    Kanban multi-agent board with zombie detection and hallucination restoration. /aim locks agent heading in the right direction throughout turns. Checkpoints v2 with actual state pruning. Gateway auto-resumes after restart. 8 P0 safety fixes: redaction on by default, Discord guild-scoped allowlists, WhatsApp stranger rejection, TOCTOU patches. Google Chat = twentieth platform. Seven i18n locales ship.

    864 commits
    588 PRs merged
    282 points closed
    295 contributors

    v0.12.0 — The Curator Launch
    Apr 30, 2026

    Autonomous background Curator grades, prunes, and consolidates the talent library. 4 new inference suppliers. Groups plugin = nineteenth platform. Native Spotify and Google Meet integrations. ComfyUI and TouchDesigner-MCP bundled. ~57% discount in TUI chilly begin time.

    v0.11.0 — The Interface Launch
    Apr 23, 2026

    Full React/Ink TUI rewrite. Native AWS Bedrock assist. 5 new inference paths: NVIDIA NIM, Arcee AI, Google Gemini CLI OAuth, Vercel ai-gateway, Step Plan. GPT-5.5 by way of Codex OAuth. QQBot = seventeenth platform. Plugin floor dramatically expanded. Dashboard with i18n (English + Chinese language).

    1,556 commits
    761 PRs merged
    29 contributors

    v0.10.0 — The Software Gateway Launch
    Apr 16, 2026

    Nous Portal subscribers get net search (Firecrawl), picture era (FAL/FLUX 2 Professional), TTS (OpenAI TTS), and browser automation (Browser Use) — zero extra API keys required.

    v0.9.0 — The In all places Launch
    Apr 13, 2026

    Android/Termux assist. iMessage by way of BlueBubbles. WeChat and WeCom adapters. Native net dashboard. Quick Mode for OpenAI and Anthropic precedence queues. Background course of monitoring by way of watch_patterns. 16 supported platforms. Deep safety hardening move.

    487 commits
    269 PRs merged
    167 points resolved

    CVE Tracker

    CVE-2026-25253 — OpenClaw

    CVSS 8.8 HIGH

    Distant exploitation of the OpenClaw gateway. Single malicious hyperlink may hand over full machine management. Patched; replace required. SecurityScorecard reported tens of 1000’s of uncovered situations.

    March 2026 Cluster — OpenClaw

    9 CVEs / 4 Days

    Mar 18–21, 2026: 9 CVEs disclosed in 4 days. One scored CVSS 9.9. Koi Safety audit of two,857 ClawHub expertise discovered 341 malicious entries (11.9%), with 335 tied to a single marketing campaign. Broader scans flagged 800+ suspicious entries.

    CVE-2026-7113 — Hermes Agent

    CVSS 3.x: 5.6 MEDIUM
    CVSS 4.0: 2.9 LOW

    Lacking authentication in webhooks endpoint (gateway/platforms/webhook.py). Impacts hermes-agent v0.8.0 solely. Excessive assault complexity; troublesome to take advantage of. Printed Apr 27, 2026. Confirm model earlier than assuming unaffected.

    v0.13.0 Safety Wave — Hermes

    8 P0 Fixes

    Tenacity launch (Could 7) closed 8 P0 safety points: redaction ON by default, Discord role-allowlists now guild-scoped, WhatsApp rejects strangers by default, TOCTOU home windows closed throughout auth.json and MCP OAuth flows.

    Select Your Stack

    Hermes Agent

    Nous Analysis — MIT License

    ▶You need an agent that improves on repeated workflows over time

    ▶Lengthy-horizon duties the place cross-session reminiscence issues

    ▶Want AWS Bedrock, NVIDIA NIM, or 200+ mannequin flexibility

    ▶Safety-first deployment (fewer CVEs, tighter defaults)

    ▶Migrating from OpenClaw (hermes claw migrate)

    ▶Serverless, SSH, or HPC deployment environments

    OpenClaw

    Impartial Basis — OpenAI Sponsor

    ▶Most messaging channel attain (50+ platforms)

    ▶Prompt entry to 44,000+ community-built expertise by way of ClawHub

    ▶Multi-agent orchestration as the first workflow

    ▶Massive group deployments needing mature, battle-tested infrastructure

    ▶You need the broader group ecosystem immediately, not compounding worth over time

    Key Takeaways

    • Hermes Agent leads OpenRouter’s international each day rankings with 224B each day tokens vs. OpenClaw’s 186B as of Could 10, 2026; OpenClaw leads cumulative all-time at 9.17T vs. 6.35T
    • OpenClaw founder Peter Steinberger joined OpenAI in February 2026; OpenClaw operates as an unbiased open-source basis with OpenAI as a sponsor
    • Hermes Agent: 114,000 GitHub stars, MIT license, 40+ built-in instruments, 20 supported messaging platforms as of v0.13.0
    • OpenClaw: 370,000 GitHub stars, 44,000+ ClawHub expertise (April 2026), 50+ messaging channels
    • Newest Hermes launch: v0.13.0 “Tenacity” (Could 7, 2026) — 864 commits, 588 PRs, 295 contributors, 8 P0 safety fixes
    • CVE-2026-25253 (CVSS 8.8) and a cluster of 9 CVEs in March 2026 stay the first OpenClaw safety threat markers; Hermes CVE-2026-7113 (CVSS 3.x: 5.6 MEDIUM) impacts model 0.8.0 solely

    Be happy to comply with us on Twitter and don’t overlook to hitch our 150k+ ML SubReddit and Subscribe to our Newsletter. Wait! are you on telegram? now you can join us on telegram as well.

    Must associate with us for selling your GitHub Repo OR Hugging Face Web page OR Product Launch OR Webinar and so forth.? Connect with us


    Michal Sutter is an information science skilled with a Grasp of Science in Information Science from the College of Padova. With a strong basis in statistical evaluation, machine studying, and information engineering, Michal excels at remodeling advanced datasets into actionable insights.



    Source link

    Naveed Ahmad

    Naveed Ahmad is a technology journalist and AI writer at ArticlesStock, covering artificial intelligence, machine learning, and emerging tech policy. Read his latest articles.

    Related Posts

    I Work in Hollywood. Everybody Who Used to Make TV Is Now Secretly Coaching AI

    11/05/2026

    Sakana AI and NVIDIA Introduce TwELL with CUDA Kernels for 20.5% Inference and 21.9% Coaching Speedup in LLMs

    11/05/2026

    A Coding Implementation to Construct Agent-Native Reminiscence Infrastructure with Memori for Persistent Multi-Person and Multi-Session LLM Functions

    11/05/2026
    Leave A Reply Cancel Reply

    Categories
    • AI
    Recent Comments
      Facebook X (Twitter) Instagram Pinterest
      © 2026 ThemeSphere. Designed by ThemeSphere.

      Type above and press Enter to search. Press Esc to cancel.