Falcon LLM

LOADING Introducing TII’s Falcon H1R 7B & H1 Arabic Making Advanced AI accessible and Available to Everyone, Everywhere Falcon H1R 7B Falcon-H1-Arabic Try our Chat pause video play video upcoming MODELS Powering the Future of Responsible AI Falcon-H1 and Falcon-E are the latest additions to the Falcon LLM family—designed to expand the possibilities of generative AI across industries and disciplines. Built with efficiency, ethical design, and global scalability in mind, these models reflect our commitment to making advanced AI accessible and impactful. From smarter language understanding to enhanced real-world applications, Falcon-H1 and Falcon-E pave the way for intelligent, inclusive, and future-ready solutions. Model Insights Falcon H1R 7B: Advanced reasoning capabilities in mathematics, coding..... Combines an attention-based and state-space architectures for efficient sequence modeling under large batch sizes and long response lengths... Learn More Model Insights Falcon-H1-Arabic: High-performance Arabic AI for the Real World..... Falcon-H1-Arabic is a complete departure from traditional large language models... Learn More Falcon-H1 Model Insights Falcon-H1: Innovative Hybrid AI Architecture Setting..... AI innovation – and its deployment in the real world – is speeding up. Until now, users had to make a trade-off between performance and efficiency. The unique hybrid architecture of Falcon-H1 (based on Transformer and Mamba architecture) gives you... Learn More Falcon Arabic Model Insights Falcon Arabic: The Future of AI in the Arabic Language Built for the Modern Standard Arabic and dialects, Falcon Arabic brings advanced language AI to the Arabic-speaking world..... Learn More Latest Updates Insights Introducing Falcon H1R 7B Falcon H1R 7B Packs Advanced Reasoning into a Compact 7 Billion Parameter Model Optimized for Speed and Efficiency TII’s Latest AI Model Outperforms Larger Rivals from Microsoft, Alibaba, and NVIDIA on Key Benchmarks View More Insights Introducing Falcon-H1 Arabic Based on new hybrid-architecture, models deliver higher accuracy while running on smaller parameter sizes Launch underscores UAE push to compete with global AI leaders in high-performance language models. View More Insights Falcon H1R 7B & H1 Arabic Falcon H1R 7B: Advanced reasoning capabilities in mathematics, coding, logic and instruction-following in a highly efficient 7B model. Falcon H1 Arabic: Falcon-H1-Arabic is a complete departure from traditional large language models. View More Insights Falcon-H1 & Falcon Arabic Falcon-H1 redefines performance and portability, outperforming Meta’s LLaMA and Alibaba’s Qwen to enable real-world AI on everyday devices and in resource-limited settings Falcon Arabic: In addition to English and European-origin languages, Falcon now integrates Arabic - expanding its reach across the Arabic-speaking world as the region’s best- performing Arabic AI model View More Insights Falcon 3 The latest updates to the Falcon 3 series introduce exceptional multimodal functionalities, enabling the models to process not only text but also images, and – for the first time in the Falcon series – video and audio data. Falcon 3 can run on light infrastructures, even laptops, without sacrificing performance The Falcon 3 ecosystem contains four scalable models tailored for diverse applications Falcon 3 supports several languages and is optimized for resource efficiency This latest iteration of TII’s open-source large language model series aims to democratize access to AI View More Insights Introducing Falcon Mamba 7B Falcon Mamba 7B is the first open source released State Space Language Model (SSLM), a new revolutionary architecture for Falcon models Falcon Mamba 7B is the no. 1 globally performing open source SSLM in the world, as independently verified by Hugging Face SSLMs have a low memory cost and don’t require additional memory to generate arbitrary long blocks of text Falcon Mamba 7B also outperforms traditional transformer architecture models such as Meta’s Llama 3.1 8B and Mistral’s 7B New model reflects the innovation and pioneering approach of Abu Dhabi in AI research and development read article Insights Falcon 2 Open-Source, Multilingual, and Multimodal – and is only AI Model with Vision-to-Language Capabilities. New Falcon 2 11B Outperforms Meta’s Llama 3 8B, and Performs on par with leading Google Gemma 7B Model, as Independently Verified by Hugging Face Leaderboard. Next up, we’re looking to add 'Mixture of Experts' to enhance Falcon 2’s capabilities even further. read article Insights UAE’s Technology Innovation Institute Launches ‘Falcon Foundation’ to Champion Open-sourcing of Generative AI Models The Technology Innovation Institute (TII), a leading global scientific research center and the applied research pillar of Abu Dhabi’s Advanced Technology Research Council (ATRC), today announced the launch of the ‘Falcon Foundation’. The non-profit Falcon Foundation is dedicated to advancing the development of open-source Generative AI models and to building sustainable ecosystems around open-source projects that accelerate technology development... read article Insights Falcon 180B Falcon 180B is a super-powerful language model with 180 billion parameters, trained on 3.5 trillion tokens. It's currently at the top of the Hugging Face Leaderboard for pre-trained Open Large Language Models and is available for both research and commercial use.. read article Insights Falcon 40B Trained on 40 billion parameters and one trillion tokens, Falcon 40B was the first home-grown open source, large language model (LLM) to be released with weights for research and commercial use. read article Call for Proposals AI71, a pioneering AI company launched by Abu Dhabi's Advanced Technology Research Council (ATRC) and VentureOne, announced a global Call for Proposals, inviting developers to submit exceptional ideas and use cases that they would like to bring to life with Falcon 10B. The best proposals will be awarded access to the necessary compute to work on their idea. Discover More Falcon LLM Hackfest: Wings of Innovation Challenge Join the Technology Innovation Institute (TII) in the Falcon LLM Hackfest: Wings of Innovation Challenge, an exciting opportunity to showcase your skills and shape the future of AI with Falcon 10B! Explore Hackfest Falcon LLMs and The Future Initiative to Convene Leading University Researchers and Global AI Practitioners to Enhance TII’s Falcon LLM Series and Champion the Open-source Movement The world’s leading university researchers and global AI practitioners can submit their proposals for how best to use Falcon for focused problem-solving challenges. The best submissions will be rewarded with compute and the opportunity to bring their ideas to life. Discover Falcon LLMs Program Our Research Making Generative AI Models Accessible Generative AI models are enabling us to create innovative pathways to an exciting future of possibilities – where the only limits are of the imagination. Our team of AI researchers, scientists, and engineers are collaborating to achieve innovative outcomes – from AI theory to AI technologies, and are sharing their efforts with the wider AI community. Datasets At the heart of our cutting-edge Language Models (LLMs) lies a robust and diverse collection of datasets. Datasets are the lifeblood of our LLM models and fortify it with a wealth of linguistic knowledge and contextual understanding. learn more Falcon Falcon LLM is a generative large language model (LLM) that helps advance applications and use cases to future-proof our world. Today the Falcon‑H1‑Tiny‑R, Falcon H1R 7B, Falcon-H1-Arabic, Falcon Arabic, Falcon-E, Falcon-H1, Falcon 3, Falcon Mamba 7B, Falcon 2, 180B, 40B, 7.5B, and 1.3B parameter AI models, as well as our high-quality REFINEDWEB dataset, form a suite of offerings. learn more Open Models Why Open Source and Open Access? As a key technology enabler, we firmly believe that innovation should be allowed to flourish. That is why we decided to open source or open access all our Falcon models. We launched our latest Falcon 180B LLM as an open access model for research and commercial use. Together, we can mobilize and fast-track breakthrough solutions worldwide - catalyzing innovation for outsized impact. Licensing and Security Ensuring Compliance and Safeguarding Your Data Falcon‑H1‑Tiny‑R: The Falcon‑H1‑Tiny series demonstrates that intelligence can indeed fit into small models, with specialized variants delivering multilingual understanding, strong instruction following, robust tool‑use and coding capabilities. Falcon‑H1‑Tiny‑R series redefines reasoning at scale by matching the performance of significantly larger reasoning models. Falcon‑H1‑Tiny‑R was trained exclusively on reasoning data and applied to compact architectures, unlocking powerful reasoning performance in the 0.6B and 0.09B configurations. Read more Falcon H1 Tiny R 0.6B (reasoning, post-GRPO) Falcon H1 Tiny R 0.6B (reasoning, pre-GRPO) Falcon H1 Tiny R 0.09B (reasoning) Falcon H1 Tiny specialized 90M models collection Falcon H1R 7B: Advanced reasoning capabilities in mathematics, coding, logic and instruction-following in a highly efficient 7B model. Falcon-H1-Arabic: Falcon-H1-Arabic is a complete departure from traditional large language models. Falcon Arabic: Built for the Modern Standard Arabic and dialects, Falcon Arabic brings advanced language AI to the Arabic-speaking world. Read our technical blog Falcon-H1: AI innovation – and its deployment in the real world – is speeding up. Until now, users had to make a trade-off between performance and efficiency. The unique hybrid architecture of Falcon-H1 (based on Transformer and Mamba architecture) gives you a family of AI models that delivers the best of both worlds – combining strong general-purpose comprehension with highly efficient processing. Thanks to an updated version of Maximal Update Parametrization it is possible to train the models reliably at very large scales; in fact, it makes the process of increasing model size much safer and more efficient. Read more about our technical blog Test Falcon on Oumi Falcon-E: Falcon Edge redefines what’s possible with small-scale language models. As AI moves beyond the cloud and into real-world devices, Falcon-e is built to work where infrastructure is limited, running fast and efficiently on CPUs, not GPUs. Read more Test Falcon on Oumi Falcon 3:The latest updates to the Falcon 3 series introduce exceptional multimodal functionalities, enabling the models to process not only text but also images, and – for the first time in the Falcon series – video and audio data. Falcon 3 can run on light infrastructures, even laptops, without sacrificing performance. The Falcon 3 ecosystem contains four scalable models tailored for diverse applications. Falcon 3 supports several languages and is optimized for resource efficiency. This latest iteration of TII’s open-source large language model series aims to democratize access to AI. The download of Falcon 3 is subject to our Terms & Conditions and Acceptable Use Policy Falcon Mamba 7B: Falcon Mamba 7B is the first open source released State Space Language Model (SSLM), a new revolutionary architecture for Falcon models. Falcon Mamba 7B is the no. 1 globally performing open source SSLM in the world, as independently verified by Hugging Face. SSLMs have a low memory cost and don’t require additional memory to generate arbitrary long blocks of text. Falcon Mamba 7B also outperforms traditional transformer architecture models such as Meta’s Llama 3.1 8B and Mistral’s 7B. New model reflects the innovation and pioneering approach of Abu Dhabi in AI research and development. The download of Falcon Mamba 7B is subject to our Terms & Conditions and Acceptable Use Policy Falcon 2: Open-Source, Multilingual, and Multimodal – and is only AI Model with Vision-to-Language Capabilities. New Falcon 2 11B Outperforms Meta’s Llama 3 8B, and Performs on par with leading Google Gemma 7B Model, as Independently Verified by Hugging Face Leaderboard. Next up, we’re looking to add 'Mixture of Experts' to enhance Falcon 2’s capabilities even further. The download of Falcon 2 is subject to our Terms & Conditions and Acceptable Use Policy Falcon 40B: Falcon 40B is made available under the permissive Apache 2.0 software license for researchers and commercial users alike. End-users have access to any patent covered by the software in question. Apache 2.0 ensures the security and availability of safe and powerful open-source software and establishes an effective governance model. Falcon 180B: Falcon-180B is accessible to developers through a royalty-free license, based on Apache 2.0. The licence includes restrictions on illegal or harmful use, and requires those intending to provide hosted access to the model to seek additional consent from TII. The model is free of charge to download, use and integrate into applications and end user products. Those hosting providers wishing to provide shared instances of the model or its derivatives as a managed service (for inference or fine tuning) are not covered by the proposed license, and are invited to enter into a separate license arrangement with TII. The download of Falcon 180B is subject to our Terms & Conditions and Acceptable Use Policy Falcon LLM Pioneering the Next Generation of Language Models Falcon LLMs and The Future Pioneering Language Models Shaping Tomorrow's Possibilities Generative AI models like Falcon Arabic, Falcon Edge, Falcon H1, Falcon 3, Falcon Mamba 7B, Falcon 2, Falcon 40B and Falcon 180B are illuminating pathways to the future. As newer and more powerful models continue to emerge every day, they will remain the UAE’s iconic trailblazers. We work tirelessly to ensure the transparency and accountability of our AI systems to protect privacy and data security, as well as establish mechanisms for addressing AI’s societal impact. By sharing our advancements, we facilitate breakthroughs in various domains, including healthcare, finance, education, and beyond. Together, we can leverage collective intelligence to solve complex challenges and drive innovation. Falcon LLM - Join us in innovating for a better world. Access Falcon Models Frequently Asked Questions I am a developer or a company who wants to use Falcon 180B in my application to provide enhanced features for my users. Is this allowed under the license? Yes. The license allows companies and developers to integrate Falcon 180B into their applications and services, and host it in their own or leased infrastructure. I am a developer or a company who wants to create a new application based on Falcon 180B (for example a chatbot) and then offer it as a paid service accessible through either a dedicated platform or an API. Is this allowed in the license? Yes. The license allows companies and developers to build and monetize new applications on top of Falcon 180B and host it in their own or leased infrastructure. I work for a major corporation. We plan to utilize Falcon 180B in our internal tools. Is this allowed under the license? Yes. The license allows companies and developers to use Falcon 180B within their internal tools and applications, and host it in their own or leased infrastructure. I am a hosting provider and I do provide dedicated infrastructure for developers or companies willing to host their dedicated instances of LLM models including Falcon 180B. Is this allowed under the license? Yes. The license doesn’t not prevent hosting providers from leasing hosting capacity / compute power that is used by developers or companies to host their dedicated instance of Falcon 180B. The license only restricts the provisioning of shared instances offered as an inference / finetuning API (subject to another license) I am a hosting provider and I want to host instances of Falcon 180B (base model or fine-tuned versions of the model) as a standalone shared service. Can I charge access to it through an API (inference API or finetuning API)? No. The license specifically restricts hosting providers from monetizing direct access to shared instances of Falcon 180B and its fine-tunings. This would require a specific license to be in place. Remember, these FAQs are based on the provided terms and conditions and aim to simplify the understanding of Falcon 180B's usage. For comprehensive information, always refer to the full License Agreement. Agile, Intuitive & Beneficial Back to Top --- LOADING Introducing TII’s Falcon H1R 7B & H1 Arabic Making Advanced AI accessible and Available to Everyone, Everywhere Falcon H1R 7B Falcon-H1-Arabic Try our Chat pause video play video upcoming MODELS Powering the Future of Responsible AI Falcon-H1 and Falcon-E are the latest additions to the Falcon LLM family—designed to expand the possibilities of generative AI across industries and disciplines. Built with efficiency, ethical design, and global scalability in mind, these models reflect our commitment to making advanced AI accessible and impactful. From smarter language understanding to enhanced real-world applications, Falcon-H1 and Falcon-E pave the way for intelligent, inclusive, and future-ready solutions. Model Insights Falcon H1R 7B: Advanced reasoning capabilities in mathematics, coding..... Combines an attention-based and state-space architectures for efficient sequence modeling under large batch sizes and long response lengths... Learn More Model Insights Falcon-H1-Arabic: High-performance Arabic AI for the Real World..... Falcon-H1-Arabic is a complete departure from traditional large language models... Learn More Falcon-H1 Model Insights Falcon-H1: Innovative Hybrid AI Architecture Setting..... AI innovation – and its deployment in the real world – is speeding up. Until now, users had to make a trade-off between performance and efficiency. The unique hybrid architecture of Falcon-H1 (based on Transformer and Mamba architecture) gives you... Learn More Falcon Arabic Model Insights Falcon Arabic: The Future of AI in the Arabic Language Built for the Modern Standard Arabic and dialects, Falcon Arabic brings advanced language AI to the Arabic-speaking world..... Learn More Latest Updates Insights Introducing Falcon H1R 7B Falcon H1R 7B Packs Advanced Reasoning into a Compact 7 Billion Parameter Model Optimized for Speed and Efficiency TII’s Latest AI Model Outperforms Larger Rivals from Microsoft, Alibaba, and NVIDIA on Key Benchmarks View More Insights Introducing Falcon-H1 Arabic Based on new hybrid-architecture, models deliver higher accuracy while running on smaller parameter sizes Launch underscores UAE push to compete with global AI leaders in high-performance language models. View More Insights Falcon H1R 7B & H1 Arabic Falcon H1R 7B: Advanced reasoning capabilities in mathematics, coding, logic and instruction-following in a highly efficient 7B model. Falcon H1 Arabic: Falcon-H1-Arabic is a complete departure from traditional large language models. View More Insights Falcon-H1 & Falcon Arabic Falcon-H1 redefines performance and portability, outperforming Meta’s LLaMA and Alibaba’s Qwen to enable real-world AI on everyday devices and in resource-limited settings Falcon Arabic: In addition to English and European-origin languages, Falcon now integrates Arabic - expanding its reach across the Arabic-speaking world as the region’s best- performing Arabic AI model View More Insights Falcon 3 The latest updates to the Falcon 3 series introduce exceptional multimodal functionalities, enabling the models to process not only text but also images, and – for the first time in the Falcon series – video and audio data. Falcon 3 can run on light infrastructures, even laptops, without sacrificing performance The Falcon 3 ecosystem contains four scalable models tailored for diverse applications Falcon 3 supports several languages and is optimized for resource efficiency This latest iteration of TII’s open-source large language model series aims to democratize access to AI View More Insights Introducing Falcon Mamba 7B Falcon Mamba 7B is the first open source released State Space Language Model (SSLM), a new revolutionary architecture for Falcon models Falcon Mamba 7B is the no. 1 globally performing open source SSLM in the world, as independently verified by Hugging Face SSLMs have a low memory cost and don’t require additional memory to generate arbitrary long blocks of text Falcon Mamba 7B also outperforms traditional transformer architecture models such as Meta’s Llama 3.1 8B and Mistral’s 7B New model reflects the innovation and pioneering approach of Abu Dhabi in AI research and development read article Insights Falcon 2 Open-Source, Multilingual, and Multimodal – and is only AI Model with Vision-to-Language Capabilities. New Falcon 2 11B Outperforms Meta’s Llama 3 8B, and Performs on par with leading Google Gemma 7B Model, as Independently Verified by Hugging Face Leaderboard. Next up, we’re looking to add 'Mixture of Experts' to enhance Falcon 2’s capabilities even further. read article Insights UAE’s Technology Innovation Institute Launches ‘Falcon Foundation’ to Champion Open-sourcing of Generative AI Models The Technology Innovation Institute (TII), a leading global scientific research center and the applied research pillar of Abu Dhabi’s Advanced Technology Research Council (ATRC), today announced the launch of the ‘Falcon Foundation’. The non-profit Falcon Foundation is dedicated to advancing the development of open-source Generative AI models and to building sustainable ecosystems around open-source projects that accelerate technology development... read article Insights Falcon 180B Falcon 180B is a super-powerful language model with 180 billion parameters, trained on 3.5 trillion tokens. It's currently at the top of the Hugging Face Leaderboard for pre-trained Open Large Language Models and is available for both research and commercial use.. read article Insights Falcon 40B Trained on 40 billion parameters and one trillion tokens, Falcon 40B was the first home-grown open source, large language model (LLM) to be released with weights for research and commercial use. read article Call for Proposals AI71, a pioneering AI company launched by Abu Dhabi's Advanced Technology Research Council (ATRC) and VentureOne, announced a global Call for Proposals, inviting developers to submit exceptional ideas and use cases that they would like to bring to life with Falcon 10B. The best proposals will be awarded access to the necessary compute to work on their idea. Discover More Falcon LLM Hackfest: Wings of Innovation Challenge Join the Technology Innovation Institute (TII) in the Falcon LLM Hackfest: Wings of Innovation Challenge, an exciting opportunity to showcase your skills and shape the future of AI with Falcon 10B! Explore Hackfest Falcon LLMs and The Future Initiative to Convene Leading University Researchers and Global AI Practitioners to Enhance TII’s Falcon LLM Series and Champion the Open-source Movement The world’s leading university researchers and global AI practitioners can submit their proposals for how best to use Falcon for focused problem-solving challenges. The best submissions will be rewarded with compute and the opportunity to bring their ideas to life. Discover Falcon LLMs Program Our Research Making Generative AI Models Accessible Generative AI models are enabling us to create innovative pathways to an exciting future of possibilities – where the only limits are of the imagination. Our team of AI researchers, scientists, and engineers are collaborating to achieve innovative outcomes – from AI theory to AI technologies, and are sharing their efforts with the wider AI community. Datasets At the heart of our cutting-edge Language Models (LLMs) lies a robust and diverse collection of datasets. Datasets are the lifeblood of our LLM models and fortify it with a wealth of linguistic knowledge and contextual understanding. learn more Falcon Falcon LLM is a generative large language model (LLM) that helps advance applications and use cases to future-proof our world. Today the Falcon‑H1‑Tiny‑R, Falcon H1R 7B, Falcon-H1-Arabic, Falcon Arabic, Falcon-E, Falcon-H1, Falcon 3, Falcon Mamba 7B, Falcon 2, 180B, 40B, 7.5B, and 1.3B parameter AI models, as well as our high-quality REFINEDWEB dataset, form a suite of offerings. learn more Open Models Why Open Source and Open Access? As a key technology enabler, we firmly believe that innovation should be allowed to flourish. That is why we decided to open source or open access all our Falcon models. We launched our latest Falcon 180B LLM as an open access model for research and commercial use. Together, we can mobilize and fast-track breakthrough solutions worldwide - catalyzing innovation for outsized impact. Licensing and Security Ensuring Compliance and Safeguarding Your Data Falcon‑H1‑Tiny‑R: The Falcon‑H1‑Tiny series demonstrates that intelligence can indeed fit into small models, with specialized variants delivering multilingual understanding, strong instruction following, robust tool‑use and coding capabilities. Falcon‑H1‑Tiny‑R series redefines reasoning at scale by matching the performance of significantly larger reasoning models. Falcon‑H1‑Tiny‑R was trained exclusively on reasoning data and applied to compact architectures, unlocking powerful reasoning performance in the 0.6B and 0.09B configurations. Read more Falcon H1 Tiny R 0.6B (reasoning, post-GRPO) Falcon H1 Tiny R 0.6B (reasoning, pre-GRPO) Falcon H1 Tiny R 0.09B (reasoning) Falcon H1 Tiny specialized 90M models collection Falcon H1R 7B: Advanced reasoning capabilities in mathematics, coding, logic and instruction-following in a highly efficient 7B model. Falcon-H1-Arabic: Falcon-H1-Arabic is a complete departure from traditional large language models. Falcon Arabic: Built for the Modern Standard Arabic and dialects, Falcon Arabic brings advanced language AI to the Arabic-speaking world. Read our technical blog Falcon-H1: AI innovation – and its deployment in the real world – is speeding up. Until now, users had to make a trade-off between performance and efficiency. The unique hybrid architecture of Falcon-H1 (based on Transformer and Mamba architecture) gives you a family of AI models that delivers the best of both worlds – combining strong general-purpose comprehension with highly efficient processing. Thanks to an updated version of Maximal Update Parametrization it is possible to train the models reliably at very large scales; in fact, it makes the process of increasing model size much safer and more efficient. Read more about our technical blog Test Falcon on Oumi Falcon-E: Falcon Edge redefines what’s possible with small-scale language models. As AI moves beyond the cloud and into real-world devices, Falcon-e is built to work where infrastructure is limited, running fast and efficiently on CPUs, not GPUs. Read more Test Falcon on Oumi Falcon 3:The latest updates to the Falcon 3 series introduce exceptional multimodal functionalities, enabling the models to process not only text but also images, and – for the first time in the Falcon series – video and audio data. Falcon 3 can run on light infrastructures, even laptops, without sacrificing performance. The Falcon 3 ecosystem contains four scalable models tailored for diverse applications. Falcon 3 supports several languages and is optimized for resource efficiency. This latest iteration of TII’s open-source large language model series aims to democratize access to AI. The download of Falcon 3 is subject to our Terms & Conditions and Acceptable Use Policy Falcon Mamba 7B: Falcon Mamba 7B is the first open source released State Space Language Model (SSLM), a new revolutionary architecture for Falcon models. Falcon Mamba 7B is the no. 1 globally performing open source SSLM in the world, as independently verified by Hugging Face. SSLMs have a low memory cost and don’t require additional memory to generate arbitrary long blocks of text. Falcon Mamba 7B also outperforms traditional transformer architecture models such as Meta’s Llama 3.1 8B and Mistral’s 7B. New model reflects the innovation and pioneering approach of Abu Dhabi in AI research and development. The download of Falcon Mamba 7B is subject to our Terms & Conditions and Acceptable Use Policy Falcon 2: Open-Source, Multilingual, and Multimodal – and is only AI Model with Vision-to-Language Capabilities. New Falcon 2 11B Outperforms Meta’s Llama 3 8B, and Performs on par with leading Google Gemma 7B Model, as Independently Verified by Hugging Face Leaderboard. Next up, we’re looking to add 'Mixture of Experts' to enhance Falcon 2’s capabilities even further. The download of Falcon 2 is subject to our Terms & Conditions and Acceptable Use Policy Falcon 40B: Falcon 40B is made available under the permissive Apache 2.0 software license for researchers and commercial users alike. End-users have access to any patent covered by the software in question. Apache 2.0 ensures the security and availability of safe and powerful open-source software and establishes an effective governance model. Falcon 180B: Falcon-180B is accessible to developers through a royalty-free license, based on Apache 2.0. The licence includes restrictions on illegal or harmful use, and requires those intending to provide hosted access to the model to seek additional consent from TII. The model is free of charge to download, use and integrate into applications and end user products. Those hosting providers wishing to provide shared instances of the model or its derivatives as a managed service (for inference or fine tuning) are not covered by the proposed license, and are invited to enter into a separate license arrangement with TII. The download of Falcon 180B is subject to our Terms & Conditions and Acceptable Use Policy Falcon LLM Pioneering the Next Generation of Language Models Falcon LLMs and The Future Pioneering Language Models Shaping Tomorrow's Possibilities Generative AI models like Falcon Arabic, Falcon Edge, Falcon H1, Falcon 3, Falcon Mamba 7B, Falcon 2, Falcon 40B and Falcon 180B are illuminating pathways to the future. As newer and more powerful models continue to emerge every day, they will remain the UAE’s iconic trailblazers. We work tirelessly to ensure the transparency and accountability of our AI systems to protect privacy and data security, as well as establish mechanisms for addressing AI’s societal impact. By sharing our advancements, we facilitate breakthroughs in various domains, including healthcare, finance, education, and beyond. Together, we can leverage collective intelligence to solve complex challenges and drive innovation. Falcon LLM - Join us in innovating for a better world. Access Falcon Models Frequently Asked Questions I am a developer or a company who wants to use Falcon 180B in my application to provide enhanced features for my users. Is this allowed under the license? Yes. The license allows companies and developers to integrate Falcon 180B into their applications and services, and host it in their own or leased infrastructure. I am a developer or a company who wants to create a new application based on Falcon 180B (for example a chatbot) and then offer it as a paid service accessible through either a dedicated platform or an API. Is this allowed in the license? Yes. The license allows companies and developers to build and monetize new applications on top of Falcon 180B and host it in their own or leased infrastructure. I work for a major corporation. We plan to utilize Falcon 180B in our internal tools. Is this allowed under the license? Yes. The license allows companies and developers to use Falcon 180B within their internal tools and applications, and host it in their own or leased infrastructure. I am a hosting provider and I do provide dedicated infrastructure for developers or companies willing to host their dedicated instances of LLM models including Falcon 180B. Is this allowed under the license? Yes. The license doesn’t not prevent hosting providers from leasing hosting capacity / compute power that is used by developers or companies to host their dedicated instance of Falcon 180B. The license only restricts the provisioning of shared instances offered as an inference / finetuning API (subject to another license) I am a hosting provider and I want to host instances of Falcon 180B (base model or fine-tuned versions of the model) as a standalone shared service. Can I charge access to it through an API (inference API or finetuning API)? No. The license specifically restricts hosting providers from monetizing direct access to shared instances of Falcon 180B and its fine-tunings. This would require a specific license to be in place. Remember, these FAQs are based on the provided terms and conditions and aim to simplify the understanding of Falcon 180B's usage. For comprehensive information, always refer to the full License Agreement. Agile, Intuitive & Beneficial Back to Top --- LOADING Back Our Research Try our Chat Making Generative AI Models Accessible Generative AI models are enabling us to create innovative pathways to an exciting future of possibilities – where the only limits are of the imagination. Our team of AI researchers, scientists, and engineers are collaborating to achieve innovative outcomes – from AI theory to AI technologies and are sharing their efforts with the wider AI community. Versatile, Instinctive Gainful Back to Top --- LOADING Back Falcon Models Try our Chat Falcon LLM is a generative large language model (LLM) that helps advance applications and use cases to future-proof our world. Today the Falcon‑H1‑Tiny‑R, Falcon H1R 7B, Falcon-H1-Arabic, Falcon Arabic, Falcon-E, Falcon-H1, Falcon 3, Falcon Mamba 7B, Falcon 2, 180B, 40B, 7.5B, 1.3B parameter AI models, as well as our high-quality REFINEDWEB dataset, form a suite of offerings. Falcon‑H1‑Tiny‑R The Falcon‑H1‑Tiny series demonstrates that intelligence can indeed fit into small models, with specialized variants delivering multilingual understanding, strong instruction following, robust tool‑use and coding capabilities. Falcon‑H1‑Tiny‑R series redefines reasoning at scale by matching the performance of significantly larger reasoning models. Falcon‑H1‑Tiny‑R was trained exclusively on reasoning data and applied to compact architectures, unlocking powerful reasoning performance in the 0.6B and 0.09B configurations. Falcon H1 Tiny R 0.6B (reasoning, post-GRPO) Falcon H1 Tiny R 0.6B (reasoning, pre-GRPO) Falcon H1 Tiny R 0.09B (reasoning) Read more about Falcon‑H1‑Tiny‑R model on our blog Falcon H1 Tiny specialized 90M models collection Falcon-H1-Arabic Falcon-H1-Arabic is a complete departure from traditional large language models. Built on a new hybrid Mamba-Transformer architecture, Falcon-H1-Arabic represents a significant evolution of the Falcon H1 family, combining higher accuracy with greater efficiency. Read our technical blog Falcon-H1-Arabic Falcon H1R 7B Advanced reasoning capabilities in mathematics, coding, logic and instruction-following in a highly efficient 7B model. Read our technical blog Falcon H1R 7B Falcon Arabic Built for the Modern Standard Arabic and dialects, Falcon Arabic brings advanced language AI to the Arabic-speaking world. Read our technical blog Falcon Arabic Falcon Arabic Falcon-H1 AI innovation – and its deployment in the real world – is speeding up. Until now, users had to make a trade-off between performance and efficiency. The unique hybrid architecture of Falcon-H1 (based on Transformer and Mamba architecture) gives you a family of AI models that delivers the best of both worlds – combining strong general-purpose comprehension with highly efficient processing. Thanks to an updated version of Maximal Update Parametrization it is possible to train the..... read more about our technical blog Falcon-H1 Test Falcon on Oumi Falcon-H1 Falcon-E Falcon Edge redefines what’s possible with small-scale language models. As AI moves beyond the cloud and into real-world devices, Falcon-E is built to work where infrastructure is limited, running fast and efficiently on CPUs, not GPUs. Read more about Falcon Edge model on our blog Test Falcon on Oumi Falcon 3 The latest updates to the Falcon 3 series introduce exceptional multimodal functionalities, enabling the models to process not only text but also images, and – for the first time in the Falcon series – video and audio data. Falcon 3 can run on light infrastructures, even laptops, without sacrificing performance. The Falcon 3 ecosystem contains four scalable models tailored for diverse applications. Falcon 3 supports several languages and is optimized for resource efficiency. This latest iteration of TII’s open-source large language model series aims to democratize access to AI. The download of Falcon 3 is subject to our Terms & Conditions and Acceptable Use Policy Falcon 3 Falcon Mamba 7B Falcon Mamba 7B is the first open source released State Space Language Model (SSLM), a new revolutionary architecture for Falcon models. Falcon Mamba 7B is the no. 1 globally performing open source SSLM in the world, as independently verified by Hugging Face. SSLMs have a low memory cost and don’t require additional memory to generate arbitrary long blocks of text. Falcon Mamba 7B also outperforms traditional transformer architecture models such as Meta’s Llama 3.1 8B and Mistral’s 7B. New model reflects the innovation and pioneering approach of Abu Dhabi in AI research and development. The download of Falcon Mamba 7B is subject to our Terms & Conditions and Acceptable Use Policy Falcon Mamba 7B Falcon 2 Falcon 2 11B | Falcon 2 11B VLM Open-Source, Multilingual, and Multimodal – and is only AI Model with Vision-to-Language Capabilities. New Falcon 2 11B Outperforms Meta’s Llama 3 8B, and Performs on par with leading Google Gemma 7B Model, as Independently Verified by Hugging Face Leaderboard. Next up, we’re looking to add 'Mixture of Experts' to enhance Falcon 2’s capabilities even further. The download of Falcon 2 is subject to our Terms & Conditions and Acceptable Use Policy Falcon 2 11B Falcon 2 11B VLM Available Soon Falcon 7B Falcon-7B is a 7B parameters causal decoder-only model built by TII and trained on 1,500B tokens of RefinedWeb enhanced with curated corpora. It is made available under the Apache 2.0 license. Download Falcon 7B Falcon 40B Falcon 40B was the world’s top-ranked open-source AI model when launched. Falcon has 40 billion parameters and was trained on one trillion tokens. For two months following its launch, Falcon 40B ranked #1 on Hugging Face’s leaderboard for open source large language models (LLMs). Offered completely royalty-free with weights, Falcon 40B is revolutionary and helps democratize AI and make it a more inclusive technology. The multilingual Falcon 40B LLM works well with English, German, Spanish, French, Italian, Portuguese, Polish, Dutch, Romanian, Czech, and Swedish languages. The foundational LLM serves as a versatile base model that can be fine-tuned for specific requirements or objectives. Falcon 40B launched a Call for Proposals from scientists, researchers, and innovators for inspiring use cases and applications with the most exceptional use cases to receive an investment of training computing power to work on the powerful model to shape transformative solutions. The model uses only 75 percent of GPT-3’s training compute, 40 percent of Chinchilla AI’s, and 80 percent of PaLM-62B’s. One of the core differences in the development of Falcon was the quality of the training data. The size of the pre-training data collected for Falcon 40B was nearly five trillion tokens gathered from public web crawls (~80%), research papers, legal text, news, literature, and social media conversations. Since LLMs are particularly sensitive to the data they are trained on, our team built a custom data pipeline to extract high-quality pre-training data using extensive filtering and deduplication, implemented both at the sample level and at the string level. The download of Falcon 40B is subject to our Terms & Conditions and Acceptable Use Policy Download Falcon 40B Falcon 180B Falcon 180B is a super-powerful language model with 180 billion parameters, trained on 3.5 trillion tokens. It's currently at the top of the Hugging Face Leaderboard for pre-trained Open Large Language Models and is available for both research and commercial use.. This model performs exceptionally well in various tasks like reasoning, coding, proficiency, and knowledge tests, even beating competitors like Meta's LLaMA 2. Among closed source models, it ranks just behind OpenAI's GPT 4, and performs on par with Google's PaLM 2 Large, which powers Bard, despite being half the size of the model. The download of Falcon 180B is subject to our Terms & Conditions and Acceptable Use Policy Download Falcon 180B Falcon Demo Try out Falcon on this demo platform! Falcon 180B Demo Responsive, Intuitive Rewarding Back to Top