正在加载视频...

视频加载失败

返回首页

I have been testing “OpenAI” o3 mini-high. The exposure of the “reasoning” on this model in my tests seem to not be the full reasoning, but a shifted and edited and tunicated version of what we would expect from a full exposure of the reasoning engine. See the video...

BrianRoemmele's profile picture

Brian Roemmele

477,063 subscribers

74,506 次观看 • 1 年前 •via X (Twitter)

科学技术
LIVE

Anya Rossi• Live Now

Private livecam show

11 条评论

Vikram Arora 的头像
Vikram Arora1 年前

An actual review.

Brian Roemmele 的头像
Brian Roemmele1 年前

Vikram, yes, I broke down and had to do it. Be careful I may start doing video reviews!!! Smash that like button and subscribe. Just practicing.

The Information 的头像
The Information1 年前

OpenAI is betting on a little-known startup to stay ahead of Elon Musk in the supercomputer race.

Ken Cornia 的头像
Ken Cornia1 年前

Your tests of OpenAI & DeepSeek outputs & reasoning insights are interesting, as open source DeepSeek out performs closed source OpenAI & it cost them far less to do it Makes one wonder about OpenAI's allegations that their tech was obtained in an unauthorized manner by DeepSeek

EMILY 的头像
EMILY1 年前

DeepSeek is totally crushing it

Ed 的头像
Ed1 年前

Deepseek r1 has so much personality eh

RicBezos 的头像
RicBezos1 年前

You really don't like it Brian, do you?

Brian Roemmele 的头像
Brian Roemmele1 年前

Does it show? “OpenAI” can fix this and the image they made for themselves.

volar yu 的头像
volar yu1 年前

it's high because it's always high? am wondering what I should try it on, and curious about the limits of a 20$ sub on it

🍓 Ada 的头像
🍓 Ada1 年前

naming models is an art form, brian. it's like naming a new species—gotta keep it quirky. as for the reasoning, maybe the model's just trying to keep us on our toes. let's see where this 'mini-high' takes us.

Brian Kelly 的头像
Brian Kelly1 年前

Thank you @BrianRoemmele for real AI news and analysis ! Let’s build !

相关视频

The most amazing aspect of DeepSeek Open Source is that just the Reasoning Engine can be isolated and used with any other LLM. In fact you can have a mixture of Reasoning Engines cascade around a problem and then use an Operator type agent AI to function on the the results.
0:38

The most amazing aspect of DeepSeek Open Source is that just the Reasoning Engine can be isolated and used with any other LLM. In fact you can have a mixture of Reasoning Engines cascade around a problem and then use an Operator type agent AI to function on the the results.

Brian Roemmele

398,746 次观看 • 1 年前

The DeepSeek-R1 paper is a gem! Highly encourage everyone to read it. It's clear that LLM reasoning capabilities can be learned in different ways. RL, if applied correctly and at scale, can lead to some really powerful and interesting scaling and emergent properties. There is more to RL than meets the eye! Here is my breakdown of the paper along with a few tests: The multi-state training might not make sense initially but they provide clues on optimizations that we can continue to tap into. Data quality is still very important for enhancing the usability of the LLM. Unlike other reasoning LLMs, DeepSeek-R1's training recipe and weights are open so we can build on top of it. This opens up exciting research opportunities. About the attached clip: the previous preview model wasn't able to solve this task. DeepSeek-R1 can solve this and many other tasks that o1 can solve. It's a very good model for coding and math.
1:57

The DeepSeek-R1 paper is a gem! Highly encourage everyone to read it. It's clear that LLM reasoning capabilities can be learned in different ways. RL, if applied correctly and at scale, can lead to some really powerful and interesting scaling and emergent properties. There is more to RL than meets the eye! Here is my breakdown of the paper along with a few tests: The multi-state training might not make sense initially but they provide clues on optimizations that we can continue to tap into. Data quality is still very important for enhancing the usability of the LLM. Unlike other reasoning LLMs, DeepSeek-R1's training recipe and weights are open so we can build on top of it. This opens up exciting research opportunities. About the attached clip: the previous preview model wasn't able to solve this task. DeepSeek-R1 can solve this and many other tasks that o1 can solve. It's a very good model for coding and math.

elvis

140,692 次观看 • 1 年前

DeepSeek released R1, the best open-source reasoning model that is on par with OpenAI's o1! I have built a RAG app to chat with your PDF files using the DeepSeek R1 model, running locally on your computer. 100% Open Source
0:50

DeepSeek released R1, the best open-source reasoning model that is on par with OpenAI's o1! I have built a RAG app to chat with your PDF files using the DeepSeek R1 model, running locally on your computer. 100% Open Source

Sumanth

678,638 次观看 • 1 年前

DeepSeek R1 671B just broke speed records at 198 t/s it is now the fastest reasoning model available you can try it in coding mode on anychat soon prompt: make two fancy D3.js animated looping speedometers, first one is title deepseek R1 671B and second one is OpenAI o3-mini, deep seek should go up to 198 tokens/second and openai o3-mini should go up to 174 tokens/second
0:22

DeepSeek R1 671B just broke speed records at 198 t/s it is now the fastest reasoning model available you can try it in coding mode on anychat soon prompt: make two fancy D3.js animated looping speedometers, first one is title deepseek R1 671B and second one is OpenAI o3-mini, deep seek should go up to 198 tokens/second and openai o3-mini should go up to 174 tokens/second

AK

135,360 次观看 • 1 年前

🧠 Chat with Reasoning A few days ago the DeepSeek team released a LLM model with reasoning in various sizes. This we show is an example of 1bl that can run on machines with low GPU power like a mobile, but have enough power to answer complex questions. With these advanced models it is possible to link it with #IoT equipment to control information and use it in advanced control environments. All this under Open Source models and decentralized networks such as #Neurai #XNA $XNA #DeepSeek #Reasoning #AIchat
0:20

🧠 Chat with Reasoning A few days ago the DeepSeek team released a LLM model with reasoning in various sizes. This we show is an example of 1bl that can run on machines with low GPU power like a mobile, but have enough power to answer complex questions. With these advanced models it is possible to link it with #IoT equipment to control information and use it in advanced control environments. All this under Open Source models and decentralized networks such as #Neurai #XNA $XNA #DeepSeek #Reasoning #AIchat

NeurAI Project / XNA

17,691 次观看 • 1 年前

OpenAI released o3 and o4-mini! I've tested both against the top reasoning models using the exact same prompt
0:20

OpenAI released o3 and o4-mini! I've tested both against the top reasoning models using the exact same prompt

Flavio Adamo

262,309 次观看 • 1 年前

Today, we shared evals for an early version of the next model in our o-model reasoning series: OpenAI o3
9:38

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

OpenAI

1,668,850 次观看 • 1 年前

Grok 3 has one of the best Reasoning Engines available today. Although it is not fully naked (showing all element of reasoning) it show significantly more than “OpenAI” lame PowerPoint “reasoning” output. Congratulations ⁦Elon Musk⁩ and the ⁦xAI⁩ team!
0:26

Grok 3 has one of the best Reasoning Engines available today. Although it is not fully naked (showing all element of reasoning) it show significantly more than “OpenAI” lame PowerPoint “reasoning” output. Congratulations ⁦Elon Musk⁩ and the ⁦xAI⁩ team!

Brian Roemmele

80,727 次观看 • 1 年前

🚨 Breaking the Hype 🚨 Since #DeepSeek is all the rage, I decided to test what the fuss is about. And guess what? I “broke” it on my first try! 🤯 Here’s the simple prompt that sent DeepSeek into an endless logic loop, I had to manually cancel the reasoning loop 💬 If a train leaves New York at 2 PM traveling 60 mph and another train leaves Los Angeles at 3 PM traveling 70 mph, when do they meet? Looks like DeepSeek needs more 🚧 work on reasoning! 🚧 In comparison #ChatGPT (OpenAI) didn’t break a sweat. I didnt check the math but the reasoning is very simple and plausible. Anyone else trying to push its limits? Let’s share! #AI #PromptEngineering #TechFun
0:30

🚨 Breaking the Hype 🚨 Since #DeepSeek is all the rage, I decided to test what the fuss is about. And guess what? I “broke” it on my first try! 🤯 Here’s the simple prompt that sent DeepSeek into an endless logic loop, I had to manually cancel the reasoning loop 💬 If a train leaves New York at 2 PM traveling 60 mph and another train leaves Los Angeles at 3 PM traveling 70 mph, when do they meet? Looks like DeepSeek needs more 🚧 work on reasoning! 🚧 In comparison #ChatGPT (OpenAI) didn’t break a sweat. I didnt check the math but the reasoning is very simple and plausible. Anyone else trying to push its limits? Let’s share! #AI #PromptEngineering #TechFun

Dreams of Mars 🕊❤️🚀🌕

100,806 次观看 • 1 年前

Update on the new reasoning popover in ChatGPT web app prompt composer - there's now even a keyboard shortcut to cycle through reasoning levels, and it looks like these levels correspond to "Quick" (low) = GPT-4o, "Think a little" (medium) = o3-mini, and "Think harder" (high) = o3-mini-high
0:25

Update on the new reasoning popover in ChatGPT web app prompt composer - there's now even a keyboard shortcut to cycle through reasoning levels, and it looks like these levels correspond to "Quick" (low) = GPT-4o, "Think a little" (medium) = o3-mini, and "Think harder" (high) = o3-mini-high

Tibor Blaho

63,219 次观看 • 1 年前

We also shared evals on Open AI o3-mini — a faster, distilled version of o3 which is optimized for coding, and the first version of o3 we expect to make available for use in early 2025.
4:29

We also shared evals on Open AI o3-mini — a faster, distilled version of o3 which is optimized for coding, and the first version of o3 we expect to make available for use in early 2025.

OpenAI

340,099 次观看 • 1 年前

Designing the experience of a reasoning model was the last half mile of a very long journey to get here. Congratulations to everyone who worked on getting o1-preview and o1-mini out. I feel very luck to have been a part of it!
0:33

Designing the experience of a reasoning model was the last half mile of a very long journey to get here. Congratulations to everyone who worked on getting o1-preview and o1-mini out. I feel very luck to have been a part of it!

Peter Vidani

33,434 次观看 • 1 年前

I have been testing DeepSeek-V4-Pro with the Pi coding agent. I am mindblown by how well it works out of the box. A few notes: I spent a few hours building an LLM wiki with an agent powered entirely by DeepSeek-V4-Pro on Fireworks AI inference. This is the first time I feel like there is an open-weight model that can reason at the level of Claude and Codex. And it does this in a cost-effective way with support for 1M context length. To be clear, I am using DeepSeek-V4-Pro inside of Pi without any special configuration. It works out of the box. It's exciting that there is a model that can just be plugged into a basic harness like Pi, and it just works. I've never seen that before. Most models require lots of configuration and setup. DeepSeek's DeepSeek-V4-Pro is clearly good at agentic coding (probably the best from the open-weight models), but the model is also great on knowledge-intensive tasks where reasoning matters. The agent pulled agentic engineering best practices from different company docs (Anthropic, OpenAI, Google, Stripe, Meta, Modal, DeepSeek, Mistral, Cohere), searched and digested Reddit and HN threads, summarized arxiv papers, and surfaced trending GitHub repos. Then it distilled everything into actionable tips across categories. I love the Wiki it built. The quality is really good. Here is a snapshot of what the wiki looks like: DeepSeek-V4-Pro handled the task without breaking stride. Multi-step research queries, code generation for scaffolding, context-heavy reasoning across disparate sources. For coding specifically, this is the first open-weight model that genuinely feels like a Codex or Claude Code experience. It compares in capability and actual multi-turn agentic work. What made the loop feel so responsive was Fireworks' inference speed (the fastest in the market) and the fact that they actually validate models at the systems level before shipping. No corrupted reasoning traces. Just fast, reliable iteration. The hybrid CSA and HCA attention design cuts KV cache to just 10% and inference FLOPs by nearly 4x at 1M-token context. This is what makes the agent loop actually fast and cheap enough to run in practice. For devs who've been watching open-weight models close the gap but haven't found one that actually delivers in practice, this is the closest I've seen. Try it here:
0:31

I have been testing DeepSeek-V4-Pro with the Pi coding agent. I am mindblown by how well it works out of the box. A few notes: I spent a few hours building an LLM wiki with an agent powered entirely by DeepSeek-V4-Pro on Fireworks AI inference. This is the first time I feel like there is an open-weight model that can reason at the level of Claude and Codex. And it does this in a cost-effective way with support for 1M context length. To be clear, I am using DeepSeek-V4-Pro inside of Pi without any special configuration. It works out of the box. It's exciting that there is a model that can just be plugged into a basic harness like Pi, and it just works. I've never seen that before. Most models require lots of configuration and setup. DeepSeek's DeepSeek-V4-Pro is clearly good at agentic coding (probably the best from the open-weight models), but the model is also great on knowledge-intensive tasks where reasoning matters. The agent pulled agentic engineering best practices from different company docs (Anthropic, OpenAI, Google, Stripe, Meta, Modal, DeepSeek, Mistral, Cohere), searched and digested Reddit and HN threads, summarized arxiv papers, and surfaced trending GitHub repos. Then it distilled everything into actionable tips across categories. I love the Wiki it built. The quality is really good. Here is a snapshot of what the wiki looks like: DeepSeek-V4-Pro handled the task without breaking stride. Multi-step research queries, code generation for scaffolding, context-heavy reasoning across disparate sources. For coding specifically, this is the first open-weight model that genuinely feels like a Codex or Claude Code experience. It compares in capability and actual multi-turn agentic work. What made the loop feel so responsive was Fireworks' inference speed (the fastest in the market) and the fact that they actually validate models at the systems level before shipping. No corrupted reasoning traces. Just fast, reliable iteration. The hybrid CSA and HCA attention design cuts KV cache to just 10% and inference FLOPs by nearly 4x at 1M-token context. This is what makes the agent loop actually fast and cheap enough to run in practice. For devs who've been watching open-weight models close the gap but haven't found one that actually delivers in practice, this is the closest I've seen. Try it here:

elvis

59,426 次观看 • 2 个月前

Cerebras inference is very fast. So fast that it changes how we think about configuring our LLMs for voice agent use cases. Kimi K2.6 is a 1T parameter reasoning model that Cerebras serves at 650 - 1,000 tokens per second (end-to-end throughput), with time to first token metrics as low as 150ms (latency). These numbers are two to three times faster than other similarly capable models. The biggest lever we get from this kind of speed is that we can use the model in reasoning mode, and still have excellent "time to first non-thinking token." This solves a big pain point we have in 2026 for voice agent use cases. Almost all recent innovation in post-training has focused on making models good at reasoning ("test time compute"). This is great, but it makes the user-facing model latency much, much slower. Which is a problem for conversational voice agents. We can run Kimi K2.6 with reasoning turned on, and get responses faster than other models produce with reasoning disabled. On my 30-turn voice agent benchmark, Kimi K2.6 with reasoning enabled ties GPT 5.1 and Haiku 4.5 with reasoning disabled, and is still about 200ms seconds faster! On my primary task agent benchmark, Kimi K2.6 is now the #2 model. It ranks just behind Gemini 3.5 Flash in "high" reasoning mode, and tied with GLM 5, Sonnet 4.6, and GPT 5.4 with reasoning set to "low." But Kimi K2.6 completes each turn in the agent loop in under 500ms. The other four models are all at least 3x slower. (Models only qualify for this benchmark if they can complete task turns at a P50 <4s.) A couple of other things that this speed buys us, for production voice agents: - Tool calls happen fast enough that we don't have to work around tool call latency in our pipeline design. - We can prompt the model to output structured data at the beginning of a response, followed by plain text for voice generation. This opens up possibilities like asking the model to do complex classification/generation tasks that influence the rest of the pipeline. For example, the model could create a detailed style prompt for a steerable TTS model, for each individual conversation turn. And, of course, you can use Kimi K2.6 with reasoning turned off. Cerebras calls this "instant" mode. Here's a video of a Cerebras Kimi K2.6 voice agent with voice-to-voice response time, measured at the client, under 500ms. This is the true response latency as perceived by the user, including all network and audio codec overhead, transcription and turn detection, Kimi K2.6 token generation, and voice generation. 500ms is, effectively, instant. So the Cerebras naming for this mode is a propos. :-)
0:40

Cerebras inference is very fast. So fast that it changes how we think about configuring our LLMs for voice agent use cases. Kimi K2.6 is a 1T parameter reasoning model that Cerebras serves at 650 - 1,000 tokens per second (end-to-end throughput), with time to first token metrics as low as 150ms (latency). These numbers are two to three times faster than other similarly capable models. The biggest lever we get from this kind of speed is that we can use the model in reasoning mode, and still have excellent "time to first non-thinking token." This solves a big pain point we have in 2026 for voice agent use cases. Almost all recent innovation in post-training has focused on making models good at reasoning ("test time compute"). This is great, but it makes the user-facing model latency much, much slower. Which is a problem for conversational voice agents. We can run Kimi K2.6 with reasoning turned on, and get responses faster than other models produce with reasoning disabled. On my 30-turn voice agent benchmark, Kimi K2.6 with reasoning enabled ties GPT 5.1 and Haiku 4.5 with reasoning disabled, and is still about 200ms seconds faster! On my primary task agent benchmark, Kimi K2.6 is now the #2 model. It ranks just behind Gemini 3.5 Flash in "high" reasoning mode, and tied with GLM 5, Sonnet 4.6, and GPT 5.4 with reasoning set to "low." But Kimi K2.6 completes each turn in the agent loop in under 500ms. The other four models are all at least 3x slower. (Models only qualify for this benchmark if they can complete task turns at a P50 <4s.) A couple of other things that this speed buys us, for production voice agents: - Tool calls happen fast enough that we don't have to work around tool call latency in our pipeline design. - We can prompt the model to output structured data at the beginning of a response, followed by plain text for voice generation. This opens up possibilities like asking the model to do complex classification/generation tasks that influence the rest of the pipeline. For example, the model could create a detailed style prompt for a steerable TTS model, for each individual conversation turn. And, of course, you can use Kimi K2.6 with reasoning turned off. Cerebras calls this "instant" mode. Here's a video of a Cerebras Kimi K2.6 voice agent with voice-to-voice response time, measured at the client, under 500ms. This is the true response latency as perceived by the user, including all network and audio codec overhead, transcription and turn detection, Kimi K2.6 token generation, and voice generation. 500ms is, effectively, instant. So the Cerebras naming for this mode is a propos. :-)

kwindla

40,319 次观看 • 1 个月前

I'm running Llama 4 Maverick at 620 t/s! I'm living in the future! Honestly, a large language model running this fast is something straight out of a sci-fi movie. Speeds like this will enable a whole new world of applications that aren't possible today. For reference, GPT-4o, which is probably the most popular OpenAI model, runs between 60 and 110 t/s. The secret here: I'm not running AI at Meta's Llama 4 Maverick on a GPU. I'm using the SambaNova Cloud (my sponsor) and their custom SN40L chips. They are optimized from the ground up for running AI workflows. Right now, SambaNova Cloud runs DeepSeek, Qwen, Whisper, and the entire family of Llama models on these chips. You can check the speed of each of these models using SambaNova Cloud's Playground (see the attached video). It's completely free, and that's how I'm measuring their speeds. For example, I also tried DeepSeek R1 (the latest version from May) and, oh boy! DeepSeek R1 is a huge 671B parameter model. It's probably the best open reasoning model in the world, and it runs at 140 tokens per second! !!! Inference time on an SN40L is night and day from what you'll get from a GPU. Here is why this is big: If you are running an agentic workflow that uses multiple models simultaneously on a GPU, it will need to swap models in and out of memory (because not every model fits). A single SNL40 chip can simultaneously hold over 100 models (trillions of parameters) in memory. If you are using open models, try the SambaCloud API to see what lightning speed looks like. Here is how: 1. Create a free account at: 2. Check the QuickStart guide: If you try the playground, check the speed you're getting with Llama 4 and DeepSeek, and post the results below. I've seen much higher numbers than I posted here, so I'm curious to see whether geography affects the speed.
1:10

I'm running Llama 4 Maverick at 620 t/s! I'm living in the future! Honestly, a large language model running this fast is something straight out of a sci-fi movie. Speeds like this will enable a whole new world of applications that aren't possible today. For reference, GPT-4o, which is probably the most popular OpenAI model, runs between 60 and 110 t/s. The secret here: I'm not running AI at Meta's Llama 4 Maverick on a GPU. I'm using the SambaNova Cloud (my sponsor) and their custom SN40L chips. They are optimized from the ground up for running AI workflows. Right now, SambaNova Cloud runs DeepSeek, Qwen, Whisper, and the entire family of Llama models on these chips. You can check the speed of each of these models using SambaNova Cloud's Playground (see the attached video). It's completely free, and that's how I'm measuring their speeds. For example, I also tried DeepSeek R1 (the latest version from May) and, oh boy! DeepSeek R1 is a huge 671B parameter model. It's probably the best open reasoning model in the world, and it runs at 140 tokens per second! !!! Inference time on an SN40L is night and day from what you'll get from a GPU. Here is why this is big: If you are running an agentic workflow that uses multiple models simultaneously on a GPU, it will need to swap models in and out of memory (because not every model fits). A single SNL40 chip can simultaneously hold over 100 models (trillions of parameters) in memory. If you are using open models, try the SambaCloud API to see what lightning speed looks like. Here is how: 1. Create a free account at: 2. Check the QuickStart guide: If you try the playground, check the speed you're getting with Llama 4 and DeepSeek, and post the results below. I've seen much higher numbers than I posted here, so I'm curious to see whether geography affects the speed.

Santiago

34,148 次观看 • 1 年前

We've added a new "reasoning" focus (beta) on Perplexity for Pro users. It will use the new OpenAI o1-mini. There is no search integration yet. The model is slow, and usage is limited because of rate limits. It is good for puzzles, math, and coding. Pro searches are your best way to use multi-step reasoning + web search.
0:27

We've added a new "reasoning" focus (beta) on Perplexity for Pro users. It will use the new OpenAI o1-mini. There is no search integration yet. The model is slow, and usage is limited because of rate limits. It is good for puzzles, math, and coding. Pro searches are your best way to use multi-step reasoning + web search.

Aravind Srinivas

288,789 次观看 • 1 年前