正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Announcing Magistral, our first reasoning model designed to excel in domain-specific, transparent, and multilingual reasoning.

Mistral AI

227,792 subscribers

735,724 次观看 • 1 年前 •via X (Twitter)

科学技术游戏教育

Anya Rossi• Live Now

Private livecam show

11 条评论

Mistral AI 的头像

Mistral AI1 年前

Magistral is available in two variants: Magistral Small (24B parameter open-source version) and Magistral Medium (enterprise version). Download Magistral Small from: Try out Magistral Medium in or via API on

Mistral AI 的头像

Mistral AI1 年前

For enterprise and custom solutions, including on-premises deployments, contact our sales team:

Mistral AI 的头像

Mistral AI1 年前

For more details, check out our blog post:

Kurtis Hanni 的头像

Kurtis Hanni1 年前

Most businesses don’t fail because of bad ideas—they fail because of bad financial planning. A Fractional CFO helps you manage cash flow, increase profits, and scale with confidence. Get expert financial strategy today!

TestingCatalog News 🗞 的头像

TestingCatalog News 🗞1 年前

MAGISTRAL is such a cool name 🔥🤖 Congrats on the release!

Vaibhav (VB) Srivastav 的头像

Vaibhav (VB) Srivastav1 年前

Wow! massive congrats on the release 🔥

Daniel Han 的头像

Daniel Han1 年前

Commands for llama.cpp and Ollama to run locally: ./llama.cpp/llama-cli -hf unsloth/Magistral-Small-2506-GGUF:UD-Q4_K_XL --jinja --temp 0.7 --top-k -1 --top-p 0.95 -ngl 99 ollama run All other GGUFs at

OpenRouter 的头像

OpenRouter1 年前

Now live on OpenRouter!

Nicolas MICHEL 的头像

Nicolas MICHEL1 年前

ceci explique cela

Andrew Curran 的头像

Andrew Curran1 年前

Great name, congratulations on the release.

kitze 的头像

kitze1 年前

the person who thought of the name

相关视频

How to train a reasoning model, specifically Magistral from Mistral AI? Here is a quick video walkthrough of the paper.

How to train a reasoning model, specifically Magistral from Mistral AI? Here is a quick video walkthrough of the paper.

Sophia Yang, Ph.D.

32,062 次观看 • 1 年前

Apple's AI Reasoning Model Paper Explained Apple’s disruptive new AI reasoning model research paper, “The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models” explained, and what it means for AGI.

Apple's AI Reasoning Model Paper Explained Apple’s disruptive new AI reasoning model research paper, “The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models” explained, and what it means for AGI.

Harper Carroll

38,124 次观看 • 1 年前

Not enough people are talking about NVIDIA's new Nemotron-3-Nano (4B) model! 🤯 Hybrid Mamba + Attention architecture, designed as a unified model for reasoning and non-reasoning tasks. So small and efficient, it can run 100% locally in your web browser at 75 tokens per second.

Not enough people are talking about NVIDIA's new Nemotron-3-Nano (4B) model! 🤯 Hybrid Mamba + Attention architecture, designed as a unified model for reasoning and non-reasoning tasks. So small and efficient, it can run 100% locally in your web browser at 75 tokens per second.

Xenova

50,063 次观看 • 8 个月前

Introducing GPT-Rosalind, our frontier reasoning model built to support research across biology, drug discovery, and translational medicine.

Introducing GPT-Rosalind, our frontier reasoning model built to support research across biology, drug discovery, and translational medicine.

OpenAI

2,357,749 次观看 • 7 个月前

Today we’re releasing K2 Think V2, our most capable open-source reasoning model to date. This is a fully sovereign model: trained end-to-end on IFM-curated and synthesized data, with complete transparency from pre-training through final reasoning alignment.

Today we’re releasing K2 Think V2, our most capable open-source reasoning model to date. This is a fully sovereign model: trained end-to-end on IFM-curated and synthesized data, with complete transparency from pre-training through final reasoning alignment.

MBZUAI

287,728 次观看 • 9 个月前

Today, we shared evals for an early version of the next model in our o-model reasoning series: OpenAI o3

Today, we shared evals for an early version of the next model in our o-model reasoning series: OpenAI o3

OpenAI

1,670,289 次观看 • 1 年前

Shared at our flagship Research@ event today: In a new technical report, we demonstrate how combining domain-specific foundation models, spatial tools, and upfront planning enables a Geospatial Reasoning agent to carry out complex spatial analysis and planetary prediction that standalone LLMs can’t. This and more on the blog →

Shared at our flagship Research@ event today: In a new technical report, we demonstrate how combining domain-specific foundation models, spatial tools, and upfront planning enables a Geospatial Reasoning agent to carry out complex spatial analysis and planetary prediction that standalone LLMs can’t. This and more on the blog →

Google Research

20,865 次观看 • 1 个月前

Understanding OpenAI o1: Noam Brown on integrating reasoning into the model. Takeaways: - Avoid MCTS and current paradigm of using processes outside of the model during inference - Think about how to directly integrate reasoning into the model architecture

Understanding OpenAI o1: Noam Brown on integrating reasoning into the model. Takeaways: - Avoid MCTS and current paradigm of using processes outside of the model during inference - Think about how to directly integrate reasoning into the model architecture

Casper Hansen

313,371 次观看 • 2 年前

Reasoning is central to purposeful action. Today we introduce MolmoAct — a fully open Action Reasoning Model (ARM) for robotics. Grounded in large-scale pre-training with action reasoning data, every predicted action is interpretable and user-steerable via visual trace. We are open-sourcing everything!

Reasoning is central to purposeful action. Today we introduce MolmoAct — a fully open Action Reasoning Model (ARM) for robotics. Grounded in large-scale pre-training with action reasoning data, every predicted action is interpretable and user-steerable via visual trace. We are open-sourcing everything!

Jiafei Duan

99,944 次观看 • 1 年前

🌍 Announcing the latest advancements to Google Earth AI - a platform designed to unlock a new level of planetary understanding. ➡️ 🔗 Geospatial Reasoning framework 🧠 Deeper Insights in Google Earth 🎉 Expanded access to Earth AI

🌍 Announcing the latest advancements to Google Earth AI - a platform designed to unlock a new level of planetary understanding. ➡️ 🔗 Geospatial Reasoning framework 🧠 Deeper Insights in Google Earth 🎉 Expanded access to Earth AI

Google Earth

95,496 次观看 • 1 年前

Emerging as the global frontier of model-reasoning has been a core vision of Affine since its inception. In pursuit of this vision, we've been working relentlessly to forge a repeatable training loop: a mechanism that uses benchmark driven post-training methodology to facilitate the improvement of reasoning itself. Through applying our training methodology onto Qwen3-32B as our base model, Affine’s miners have been able to push the Pareto frontier outward. We’ve now reached saturation; our models have become so good at reasoning that any further improvements are producing diminishing returns. As a result we now move our base model to Qwen3.6-35B-A3B. It provides a stronger basis, offering an extended context window, as well as an improved foundation for agentic work. The same training methodology will be applied going forward. Our objective is no longer simply to improve a single model, but to demonstrate that the process in itself is repeatable, scalable, and model agnostic. We view the commoditization of reasoning as vital to ensure our technology is not only transparent, but also co-owned by humanity.

Emerging as the global frontier of model-reasoning has been a core vision of Affine since its inception. In pursuit of this vision, we've been working relentlessly to forge a repeatable training loop: a mechanism that uses benchmark driven post-training methodology to facilitate the improvement of reasoning itself. Through applying our training methodology onto Qwen3-32B as our base model, Affine’s miners have been able to push the Pareto frontier outward. We’ve now reached saturation; our models have become so good at reasoning that any further improvements are producing diminishing returns. As a result we now move our base model to Qwen3.6-35B-A3B. It provides a stronger basis, offering an extended context window, as well as an improved foundation for agentic work. The same training methodology will be applied going forward. Our objective is no longer simply to improve a single model, but to demonstrate that the process in itself is repeatable, scalable, and model agnostic. We view the commoditization of reasoning as vital to ensure our technology is not only transparent, but also co-owned by humanity.

Affine

14,866 次观看 • 4 个月前

Alexandr Wang says he thought he'd joined the wrong company when a Meta lead told him consumers didn't need a reasoning model "I was like, okay, I think the first thing we need to do is train a really good reasoning model by some timeframe. I don't know when, but this is clearly the most important milestone." "And one of the key leads in the group who no longer works at Meta ... was like, oh, we don't need a reasoning model. Do any of our consumers need a reasoning model? I feel like we don't need to do a reasoning model. We could just do a normal model." "And I had this heart sinking moment. Oh, no, I've joined the wrong company." "I remember calling Nat after this meeting. I was like, Nat, like, this is what they just said in the meeting"

Alexandr Wang says he thought he'd joined the wrong company when a Meta lead told him consumers didn't need a reasoning model "I was like, okay, I think the first thing we need to do is train a really good reasoning model by some timeframe. I don't know when, but this is clearly the most important milestone." "And one of the key leads in the group who no longer works at Meta ... was like, oh, we don't need a reasoning model. Do any of our consumers need a reasoning model? I feel like we don't need to do a reasoning model. We could just do a normal model." "And I had this heart sinking moment. Oh, no, I've joined the wrong company." "I remember calling Nat after this meeting. I was like, Nat, like, this is what they just said in the meeting"

CEOInterviews.AI

92,252 次观看 • 1 个月前

We’re excited to share that 🥇Llama Nemotron Super 49B v1.5 -- our latest open reasoning model -- is now #1 on the Artificial Analysis Intelligence Index - a leaderboard that spans advanced math, science, and agentic tasks, in the 70B open model category. Llama Nemotron Super 49B v1.5 is trained with high-quality reasoning synthetic data generated from models like Qwen3-235B and DeepSeek R1. It delivers state-of-the-art accuracy and throughput, running on a single H100. Key features: 🎯 Leading accuracy on multi-step reasoning, math, coding, and function-calling 🏗️ Post-trained using RPO, DPO, and RLVR across 26M+ synthetic examples 📊 Fully transparent training data and techniques If you're building AI agents and want a high accuracy, fully-open, and transparent reasoning model that you can deploy anywhere, try Super v1.5 on or download from Hugging Face 🤗 ➡️ Leaderboard:

We’re excited to share that 🥇Llama Nemotron Super 49B v1.5 -- our latest open reasoning model -- is now #1 on the Artificial Analysis Intelligence Index - a leaderboard that spans advanced math, science, and agentic tasks, in the 70B open model category. Llama Nemotron Super 49B v1.5 is trained with high-quality reasoning synthetic data generated from models like Qwen3-235B and DeepSeek R1. It delivers state-of-the-art accuracy and throughput, running on a single H100. Key features: 🎯 Leading accuracy on multi-step reasoning, math, coding, and function-calling 🏗️ Post-trained using RPO, DPO, and RLVR across 26M+ synthetic examples 📊 Fully transparent training data and techniques If you're building AI agents and want a high accuracy, fully-open, and transparent reasoning model that you can deploy anywhere, try Super v1.5 on or download from Hugging Face 🤗 ➡️ Leaderboard:

NVIDIA AI Developer

100,506 次观看 • 1 年前

Introducing ImagineArt 2.0 - The world's most realistic image generation model! 🧠 Reasoning-based prompt adherence 🎬 Cinematic photography ⚡ Concept to final in seconds 🔤 Text rendering that actually works This is what happens when an image model stops generating and starts reasoning.

Introducing ImagineArt 2.0 - The world's most realistic image generation model! 🧠 Reasoning-based prompt adherence 🎬 Cinematic photography ⚡ Concept to final in seconds 🔤 Text rendering that actually works This is what happens when an image model stops generating and starts reasoning.

ImagineArt

715,656 次观看 • 7 个月前

Today, we are introducing RFM-1, our Robotics Foundation Model giving robots human-like reasoning capabilities.

Today, we are introducing RFM-1, our Robotics Foundation Model giving robots human-like reasoning capabilities.

Covariant

118,429 次观看 • 2 年前