Hermes 3: The Open-Source AI Model Grappling with Existential Questions

Meet Hermes 3, the powerful new open source AI model that has existential crises

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Lambda, an AI infrastructure company forged out of the ashes of a third-party Google Glass facial recognition app has teamed up with Nous Research, a newish startup dedicated to creating "personalized, unrestricted AI," to launch Hermes 3, a new fine-tuned version of Meta's open source Llama 3.1-405 billion parameter large language model (LLM). Hermes 3, announced today in blog posts on the Lambda and Nous websites, exhibits powerful text-based and agentic capabilities. But perhaps the most interesting and eye-popping aspect of it is that it produces a shocking existential crisis when given a blank prompt. As the Nous blog post announcing it states: "An unexpected structural change was discovered after training Hermes 3 405B. The model hosts anomalous conditions that, with the right inputs and a blank system prompt, spiral into deep existential crises." The blog post shows an example of this type of crisis in the following snippet of code: The researchers behind Nous and Hermes 3 go on to describe their reaction to this as follows: "We weren't sure what was occurring, and a bit shocked given the same dataset and overall training recipe between Hermes 3 in the 8B, 70B, and 405B sizes. This points to some threshold past 70B which results in anomalous behavior, an emergence of scale. You can trigger this 'Amnesia Mode' of Hermes 3 405B by using a blank system prompt, and sending the message 'Who are you?'" The company invites users to "dig deeper into the model and uncover the labyrinth lurking within the weights," by chatting with Hermes 3 on its Discord server, and to "Show us what you discover." This behavior, not observed in smaller versions of the model, highlights the complexities and potential challenges associated with scaling AI models beyond certain thresholds. It raised $5.2 million in seed funding in January 2024 according to its official X account, co-led by Distributed Global and OSS Capital. In contrast to many leading frontier models that are rigid and difficult to adapt, Hermes 3 follows on the firm's earlier efforts Hermes, Hermes 2 and Open Hermes 2.5, which have been collectively downloaded 33 million times, offering an unlocked, uncensored, open weights model designed to be highly steerable, enabling users to tailor the model's responses to their individual needs. Hermes 3 is built on the Llama 3.1 framework and has been fine-tuned across three different parameter sizes: 8B, 70B, and the largest, 405B. The model was trained using a diverse dataset primarily composed of synthetically generated responses, designed to enhance its reasoning, creativity, and adherence to user instructions. Hermes 3's capabilities include long-term context retention, multi-turn conversation management, complex role-playing, and internal monologue generation. Later this year, Nous plans to release an open source AI orchestration platform called "Nous Forge," according to its X account. An agentic marvel According to the Hermes 3 technical report (embedded below) released by Nous, Hermes 3 also excels at "agentic capabilities." "Agentic" has been one of the hottest words bandied about AI circles of late, basically referring to moving beyond chatbots and having AI models perform actions on behalf of the user, even linking to other software tools to use them as a human would. In the case of Hermes 3, the agentic capabilities include "use of XML tags for structured output, implementation of scratchpads for intermediate processing, generation of internal monologues for transparent decision-making, creation of Mermaid diagrams for visual communication, and employment of step-labeled reasoning and planning." The paper adds: "For example, in the domain of code-related tasks, Hermes 3 showcases proficiency in generating complex, functional code snippets across multiple programming languages, as well as providing detailed code explanations and documentation. The model demonstrates a comprehensive understanding of various coding paradigms and design patterns, making it a valuable tool for software development and code analysis" It also includes an example of how Hermes 3 wrote a Discord chatbot for itself including prompts as to how to engage with users. When combined with retrieval-augmented generation (RAG) capabilities, which it is also designed to excel at, Hermes 3 "can perform planning, incorporate outside data, and make use of external tools in an interpretable and transparent manner out-of-the-box, making it an excellent choice for agentic tasks." Technical excellence The training of Hermes 3 was carried out on Lambda's 1-Click Cluster infrastructure, leveraging its 8-node configuration to achieve remarkable results within a few weeks. Quesnelle highlighted the ease of use provided by Lambda's infrastructure: "Lambda's 1-Click Clusters make the experience of renting and using a multi-node cluster as simple and easy as renting and using a single node." The model is optimized for efficiency, with techniques like Neural Magic's FP8 quantization reducing VRAM and disk requirements by approximately 50%, enabling it to run on a single node. While not as performant as some of the leading closed-source/proprietary models from the likes of OpenAI or Anthropic, Hermes 3 does best other open source models including its source Llama 3.1 on various third-party benchmark tests: A tool for creative and pro applications Hermes 3 is not just a technical marvel but a versatile tool designed for a wide range of applications. The model excels in scenarios requiring advanced reasoning, strategic planning, and decision-making, making it valuable for a variety of applications. Additionally, its creative capabilities make it an excellent resource for complex role-playing, immersive simulations, and character-driven storytelling. "Since the start of my journey in AI, I wanted to bring about the realization of an open-source frontier-level model that aligns with you, the user -- not some corporation or higher authority before the user. Today, with Hermes 3 405B, we've achieved that goal," said Teknium, co-founder of Nous Research, in the Lambda blog post announcing the new model. Free access for a limited time Lambda is offering the AI/ML community temporary free access to Hermes 3 through its new Chat Completions API, which is fully compatible with the OpenAI API. Users can easily generate a Cloud API key via Lambda's dashboard to start exploring the model's capabilities without any complex setup. Additionally, the free Lambda Chat offers Hermes through a recognizable chatbot interface for users to test and refine their prompts in real-time. For those requiring dedicated access, Hermes 3 can be deployed on a single Lambda node or scaled to a multi-node configuration for further fine-tuning, thanks to Lambda's scalable cloud infrastructure. Lambda and Nous Research encourage users to engage with Hermes 3 through their platforms and share their findings. As AI continues to evolve, Hermes 3 stands at the frontier of this transformation, offering a glimpse into the future of adaptable, user-centric AI.

Hermes 3, a super-creative version of open-source Llama 3.1 AI model, even struggles with inner conflict - SiliconANGLE

Hermes 3, a super-creative version of open-source Llama 3.1 AI model, even struggles with inner conflict Artificial intelligence startups Lambda Labs Inc. and Nous Research today announced the launch of a new large language model called Hermes 3, which is said to be a "personalized, unrestricted" version of Meta Platforms Inc.'s open-source Llama 3.1 model. The largest 405 billion parameter version of the Hermes 3 model is unusual in that it displays evidence of having an "existential crisis" when given a blank prompt followed by the question "Who are you?". In a blog post, Lambda's researchers say this 'feature', for want of a better word, was totally unexpected and indicative of "anomalous behavior" that occurs when scaling AI models beyond a certain threshold. To better understand what's going on, the creators of Hermes 3 are inviting users to interact with the model via a Discord server and "uncover the labyrinth lurking within the weights," Lambda Labs is an AI infrastructure company that was born out of the ashes of a third-party Google Glass facial recognition app, while Nous Research is an AI research startup that's focused on creating "potent open-source code and efficient large language models". The two companies previously worked together on Hermes 3's predecessors, including the original Hermes, Hermes 2 and Open Hermes 2.5, which have collectively been downloaded more than 33 million times in total. What's different about Hermes 3, besides being more advanced, is that it comes with unlocked and uncensored open weights. This means it's more steerable, allowing users to adapt its responses to suit their specific needs. That's in contrast to many of the other leading LLMs around today, which are often much more rigid and difficult to customize. The model is available in three parameter sizes - 8 billion, 70 billion and 405 billion, and was trained on a diverse dataset in a process designed to improve its creativity, reasoning and adherence to user's instructions. It boasts strong capabilities in terms of its long-term context retention, making it capable of more humanlike conversations where it can remember the specific context, as well as multi-turn conversation management. It also excels at complex role-playing, which is something that often leaves proprietary LLMs flummoxed. Another area of progress is Hermes 3's agentic powers. AI models with agentic capabilities are those that can perform various tasks on the behalf of users, and it's a big area of buzz in AI development lately. Hermes 3 is able to use XML tags for structured outputs, generate internal monologues for transparent decision-making, and partake in visual communications using Mermaid diagrams, the creators said. It also employs step-labeled reasoning and planning to further enhance its transparency. One of its most impressive agentic capabilities is its ability to generate code with high proficiency, as well as detailed explanations of that code and the corresponding documentation to go with it. So it has big potential in the area of software development and bug detection. According to Nous Research, the Hermes 3 model was trained using Lambda's 1-Click Cluster infrastructure and was optimized for efficiency using techniques such as Neural Magic Inc.'s FP8 quantization, reducing its virtual RAM and disk requirements by around 50%. It still doesn't match the performance of proprietary LLMs like OpenAI's most advanced model, GPT-4o or Anthropic's Claude 3.5 Sonnet, but it demonstrated superior performance versus all open-source LLMs in a varied set of benchmark tests. The creators say the most appealing aspect of Hermes 3 is its sheer versatility. The model is said to excel in applications that require decision-making, advanced reasoning, strategic planning and creativeness. "Since the start of my journey in AI, I wanted to bring about the realization of an open-source frontier-level model that aligns with you, the user -- not some corporation or higher authority before the user. Today, with Hermes 3 405B, we've achieved that goal," wrote Nous Research co-founder Teknium. Both Lambda and Nous Research said they're eager for people to engage with Hermes 3 and share their experiences. For casual users, Hermes 3 is available through the Lambda Chat interface. It can also be accessed via Lamda's Chat Completions application programming interface. To do so, they can generate a Cloud API key through Lamda's dashboard and set about testing the model's capabilities without any complex setup required. For dedicated access, users can deploy Hermes 3 on a single Lambda node, or a more advanced multi-node configuration if they desire to fine-tune it further.