• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI
    Announcement

    Reflection AI introduces Beam, a 501B-parameter open model

    Reflection says Beam uses 23B active parameters and plans to publish its full weights later in October 2026.

    Nathan LambertNL
    Ofir PressOP
    Graham NeubigGN
    64 Sources, ,

    TLDR

    Reflection AI introduced Beam as its first model, with 501 billion total parameters and 23 billion active. The company says it pretrained Beam on 24 trillion tokens in four weeks. Full weights, a model card and a technical report are planned later in October 2026; team member Alex Polozov says the weights will use Apache 2.0.

    Combined views

    682.8K

    64 Sources, first seen 3h ago

    Combined views

    682.8K

    64 Sources, first seen 3h ago

    7.5K likes

    Useful links

    Sources Podcast · YouTube

    Reflection’s founders on building a DeepSeek of the West

    Washington Post Live · YouTube

    Reflection CEO Misha Laskin on open source AI models as a national security imperative
    3h ago
    first seen 3h ago
    7.5K likes
    523 comments
    1.8K saves
    882 reposts
    523 comments
    1.8K saves
    882 reposts

    Reflection AI has introduced Beam, its first model, with 501 billion total parameters and 23 billion active parameters. The company describes it as an agentic open model trained end to end from scratch.

    Featured Source

    The announcement comes before the full weights release. Reflection's launch post says the weights are due later in October 2026, meaning the complete weights were not yet public when the company introduced Beam.

    Four weeks of pretraining

    Reflection says it pretrained Beam in four weeks on 24 trillion tokens. The company credits work on mixture-of-experts stability, large-scale data curation and deduplication for building the model's foundation.

    Sentiment

    Positive61.8%38.2%Negative

    Summary

    Positive accounts welcomed Reflection's Beam model launch for its inference efficiency gains and team achievements, while negative replies questioned the marketing around its scale and compared it unfavorably to existing models.

    Based on 62 sentiment-bearing replies from 59 accounts across 4 conversations.

    Related Videos

    • Reflection’s founders on building a DeepSeek of the WestSources Podcast · YouTube
    • Reflection CEO Misha Laskin on open source AI models as a national security imperativeWashington Post Live · YouTube
    Today's Rank

    #1

    Today's Rank

    #1

    Ioannis Antonoglou, a member of the Reflection team, said pretraining began in July. He described Beam as a first milestone toward the company's goal of making open intelligence broadly accessible, and said Reflection is already training a larger successor.

    What Reflection plans to release

    Antonoglou said the company plans to publish Beam's full weights, model card and technical report later in October. Team member Alex Polozov added that the weights are planned under the Apache 2.0 license, alongside a detailed technical report and integrations with open-source partners.

    Until those materials are public, the model specifications and performance descriptions remain claims from Reflection and its team.

    64 Sources

    Deirdre Bosa@dee_bosaBIG news in AI: Reflection has finally unveiled its much anticipated open weight model, Beam... aiming to be the American answer to China’s open model lead. As the AI race expands beyond OpenAI and Anthropic, open weight models are surging with developers and moving quickly into the enterprise. China has dominated with GLM, Kimi and DeepSeek. If Beam can deliver on both capability and efficiency, America may finally have a real contender https://reflection.ai/blog/introducing-beam3h
    Reflection@reflection_aiIntroducing Beam: a highly efficient agentic open model with 501B total parameters and 23B active. - Frontier reasoning efficiency - Advances the Western open frontier on coding & agentic tasks - Trained end-to-end from scratch Full weights release this month. Learn more about Beam: http://reflection.ai/beam3h
    Ioannis Antonoglou@real_ioannisToday, we’re introducing Beam, Reflection’s first model. A year ago, we set out to build a frontier open model before we had much of the team or infrastructure required to do it. Since then, Reflection has grown 10x. We assembled an extraordinary technical team of researchers and engineers I feel privileged to call my colleagues, alongside a world-class group across recruiting, business, policy, operations, and the rest of the company whose work made all of this possible. At the same time, we built the science, data, software, and infrastructure behind the model itself. In July, we started pretraining Beam and weeks later, we had a 500B-parameter model at the frontier of Western open intelligence—particularly strong at agentic coding, with remarkable reasoning efficiency. What makes me proudest, though, is our team. The relentless focus, speed of execution and the willingness to own and solve whatever problem was in front of us. And, most importantly, the commitment of this team to our mission. Beam is our first milestone toward our mission of building open intelligence and making it accessible to everyone. We’ll release Beam’s weights, model card, and tech report later this month. And we’re already training the next, larger model. Over the past year, we built the team and the engine to propel us to the frontier and that’s what makes me so excited about what comes next.3h
    🇺🇦 Alex Polozov@alexpolozovI am so proud of Beam. Beam is Reflection's first open-weight model. It's the result of a ~year of hard work, ingenuity, and camaraderie of one of the most incredible teams I've ever seen in AI 🔥 I joined Reflection last November. Pretraining in particular was so nascent – only 5-10 people and a bit of code! We didn't yet have a crawler, or a resilient GPU infra, or a deduplication pipeline, or in-house evals, or even sufficiently big MoEs. All that started in earnest in 2026, in one of the most satisfying sprints I've ever had the pleasure to live through. What we nailed, I think, was a perfect blend of scientific rigor, startup intensity, team iteration, and ambition. We did not YOLO decisions in pretraining, that tends to blow up in your face when you scale up 🙂 But we also made sure to move fast and play to our strengths as a nimble, high-ownership, high-trust team. The resulting system of systematic iterations and fast cross-team back-and-forths really paid off. The midtraining, RL, and post-training teams really really cooked here on top of the pretrained base. The announcement below details the sheer scale and payoff of that investment, I'll let the team cover it, it's remarkable. On many agentic coding capabilities the model kept going up without any sign of a plateau. Feel free to read more in the announcement today – but later this month we will also drop the full weights release under Apache 2.0, a detailed tech report with the cool LLM science, and partner integrations with the OSS ecosystem. It's an open model after all. I strongly believe open intelligence is the future of AI, this is why I'm here. And I encourage y'all to join us on the open side 😉2h
    Andrew Carr 🤸@andrew_n_carrLet's go!!!! super excited for this release2h
    Sriram Krishnan@sriramkcongratulations to @reflection_ai on the launch of Beam. It is critical for the U.S. to have leading open models to win the AI race. I know @MishaLaskin and team have been working on this for a while and very excited to see them launch this.2h
    Tanishq Mathew Abraham, Ph.D.@iScienceLuvrCongrats to Reflection for launching their first model! It took a while but they got there😄 Looks like a GLM-5.2-level model, perhaps the best American open-source model? Weights will be released soon... Overall, it's great to have another strong open-source player in the US2h
    Chubby♨️@kimmonismusReflection has announced Beam: a 501B-parameter model with 23B active, with full weights scheduled for release this month. Frontier capabilities, and holy is this model efficient! Trained from scratch, Beam targets coding, reasoning and agentic tasks. Reflection claims 3–4x higher inference efficiency than GLM 5.2 and over 4x higher efficiency than leading Western open models in its class. 24T pretraining tokens in four weeks, followed by four weeks of reinforcement learning on 10,500 GB300s. Terminal Bench v2.1 and 80.9 on SWE-bench Verified. But again: look at the efficency, eypecially in Terminal bench v2.1! Congrats on that launch. More western open models to come.2h
    Behrooz Ghorbani@_ghorbaniToday, we’re introducing Beam, @reflection_ai's first model. We started the high-compute RL team in February. Looking back at these past few months, it’s remarkable how far we’ve come. Making RL work at this scale took careful science, scalable infrastructure, and a strong foundation from mid-training. Seeing that work come together to keep learning stable as we pushed the scale and complexity of the run has been incredibly rewarding. What has made this especially meaningful is the people: their commitment and curiosity, the way they supported one another through difficult stretches, and the joy of figuring things out together. To everyone who helped bring Beam to life, across RL science, infrastructure, mid-training, pretraining, and our close collaborators throughout Reflection: thank you. There is so much of your hard work and care in this model. Building and leading this team has been an honor and one of the most rewarding experiences of my career. I’ve learned so much from the people around me, and I’m proud of what we’ve built together. @MishaLaskin and @real_ioannis, thank you for your trust and support throughout this journey. This is our first model, and there’s so much more ahead. I’m excited to share Beam with the community, learn from your feedback, and continue building better models together.2h
    Stephanie Zhan@stephzhanRT @real_ioannis: Today, we’re introducing Beam, Reflection’s first model. A year ago, we set out to build a frontier open model before we…2h

    Sentiment

    Positive61.8%38.2%Negative

    Summary

    Related Videos

    Positive accounts welcomed Reflection's Beam model launch for its inference efficiency gains and team achievements, while negative replies questioned the marketing around its scale and compared it unfavorably to existing models.

    Based on 62 sentiment-bearing replies from 59 accounts across 4 conversations.

    Reflection’s founders on building a DeepSeek of the WestSources Podcast · YouTube
  • Reflection CEO Misha Laskin on open source AI models as a national security imperativeWashington Post Live · YouTube

    • 64 Sources

    Deirdre Bosa@dee_bosaBIG news in AI: Reflection has finally unveiled its much anticipated open weight model, Beam... aiming to be the American answer to China’s open model lead. As the AI race expands beyond OpenAI and Anthropic, open weight models are surging with developers and moving quickly into the enterprise. China has dominated with GLM, Kimi and DeepSeek. If Beam can deliver on both capability and efficiency, America may finally have a real contender https://reflection.ai/blog/introducing-beam3h
    Reflection@reflection_aiIntroducing Beam: a highly efficient agentic open model with 501B total parameters and 23B active. - Frontier reasoning efficiency - Advances the Western open frontier on coding & agentic tasks - Trained end-to-end from scratch Full weights release this month. Learn more about Beam: http://reflection.ai/beam3h
    Ioannis Antonoglou@real_ioannisToday, we’re introducing Beam, Reflection’s first model. A year ago, we set out to build a frontier open model before we had much of the team or infrastructure required to do it. Since then, Reflection has grown 10x. We assembled an extraordinary technical team of researchers and engineers I feel privileged to call my colleagues, alongside a world-class group across recruiting, business, policy, operations, and the rest of the company whose work made all of this possible. At the same time, we built the science, data, software, and infrastructure behind the model itself. In July, we started pretraining Beam and weeks later, we had a 500B-parameter model at the frontier of Western open intelligence—particularly strong at agentic coding, with remarkable reasoning efficiency. What makes me proudest, though, is our team. The relentless focus, speed of execution and the willingness to own and solve whatever problem was in front of us. And, most importantly, the commitment of this team to our mission. Beam is our first milestone toward our mission of building open intelligence and making it accessible to everyone. We’ll release Beam’s weights, model card, and tech report later this month. And we’re already training the next, larger model. Over the past year, we built the team and the engine to propel us to the frontier and that’s what makes me so excited about what comes next.3h
    🇺🇦 Alex Polozov@alexpolozovI am so proud of Beam. Beam is Reflection's first open-weight model. It's the result of a ~year of hard work, ingenuity, and camaraderie of one of the most incredible teams I've ever seen in AI 🔥 I joined Reflection last November. Pretraining in particular was so nascent – only 5-10 people and a bit of code! We didn't yet have a crawler, or a resilient GPU infra, or a deduplication pipeline, or in-house evals, or even sufficiently big MoEs. All that started in earnest in 2026, in one of the most satisfying sprints I've ever had the pleasure to live through. What we nailed, I think, was a perfect blend of scientific rigor, startup intensity, team iteration, and ambition. We did not YOLO decisions in pretraining, that tends to blow up in your face when you scale up 🙂 But we also made sure to move fast and play to our strengths as a nimble, high-ownership, high-trust team. The resulting system of systematic iterations and fast cross-team back-and-forths really paid off. The midtraining, RL, and post-training teams really really cooked here on top of the pretrained base. The announcement below details the sheer scale and payoff of that investment, I'll let the team cover it, it's remarkable. On many agentic coding capabilities the model kept going up without any sign of a plateau. Feel free to read more in the announcement today – but later this month we will also drop the full weights release under Apache 2.0, a detailed tech report with the cool LLM science, and partner integrations with the OSS ecosystem. It's an open model after all. I strongly believe open intelligence is the future of AI, this is why I'm here. And I encourage y'all to join us on the open side 😉2h
    Andrew Carr 🤸@andrew_n_carrLet's go!!!! super excited for this release2h
    Sriram Krishnan@sriramkcongratulations to @reflection_ai on the launch of Beam. It is critical for the U.S. to have leading open models to win the AI race. I know @MishaLaskin and team have been working on this for a while and very excited to see them launch this.2h
    Tanishq Mathew Abraham, Ph.D.@iScienceLuvrCongrats to Reflection for launching their first model! It took a while but they got there😄 Looks like a GLM-5.2-level model, perhaps the best American open-source model? Weights will be released soon... Overall, it's great to have another strong open-source player in the US2h
    Chubby♨️@kimmonismusReflection has announced Beam: a 501B-parameter model with 23B active, with full weights scheduled for release this month. Frontier capabilities, and holy is this model efficient! Trained from scratch, Beam targets coding, reasoning and agentic tasks. Reflection claims 3–4x higher inference efficiency than GLM 5.2 and over 4x higher efficiency than leading Western open models in its class. 24T pretraining tokens in four weeks, followed by four weeks of reinforcement learning on 10,500 GB300s. Terminal Bench v2.1 and 80.9 on SWE-bench Verified. But again: look at the efficency, eypecially in Terminal bench v2.1! Congrats on that launch. More western open models to come.2h
    Behrooz Ghorbani@_ghorbaniToday, we’re introducing Beam, @reflection_ai's first model. We started the high-compute RL team in February. Looking back at these past few months, it’s remarkable how far we’ve come. Making RL work at this scale took careful science, scalable infrastructure, and a strong foundation from mid-training. Seeing that work come together to keep learning stable as we pushed the scale and complexity of the run has been incredibly rewarding. What has made this especially meaningful is the people: their commitment and curiosity, the way they supported one another through difficult stretches, and the joy of figuring things out together. To everyone who helped bring Beam to life, across RL science, infrastructure, mid-training, pretraining, and our close collaborators throughout Reflection: thank you. There is so much of your hard work and care in this model. Building and leading this team has been an honor and one of the most rewarding experiences of my career. I’ve learned so much from the people around me, and I’m proud of what we’ve built together. @MishaLaskin and @real_ioannis, thank you for your trust and support throughout this journey. This is our first model, and there’s so much more ahead. I’m excited to share Beam with the community, learn from your feedback, and continue building better models together.2h
    Stephanie Zhan@stephzhanRT @real_ioannis: Today, we’re introducing Beam, Reflection’s first model. A year ago, we set out to build a frontier open model before we…2h