Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

OpenAI Decision crushed by Microsoft-Decision-1 at Space Invaders 👾 Decision-1 decided 3.5x faster so it scored 3,320 and cleared 4 waves while OpenAI's decision API scored 1,300 taking over 300ms per decision on average Run decision models ->

atomic_chat_hq's profile picture

atomic.chat

18,917 subscribers

91,076 Aufrufe • vor 9 Tagen •via X (Twitter)

Wissenschaft & Technologie
LIVE

Anya Rossi• Live Now

Private livecam show

10 Kommentare

Profilbild von Anderson
Andersonvor 9 Tagen

@Microsoft beating a benchmark at space invaders is the least impressive flex in AI right now

Profilbild von L3bao
L3baovor 9 Tagen

@Microsoft But wouldnt the fire rate play into this as well? Is it manually deciding to fire or is it a fixed rate? This demo really seems to favor the Microsoft model.

Profilbild von Jabez
Jabezvor 9 Tagen

@Microsoft Your logo is Claude but black and white? 🤔

Profilbild von Cuth
Cuthvor 9 Tagen

@Microsoft adding to the benchmark

Profilbild von Matin Urdu
Matin Urduvor 9 Tagen

@Microsoft Vision only?

Profilbild von Georgy GLuhov
Georgy GLuhovvor 9 Tagen

@Microsoft What about Vega?

Profilbild von Baha 🚲
Baha 🚲vor 9 Tagen

@Microsoft which decision model is the best all over?

Profilbild von Roland Brüggemann | EVELIQ Trace
Roland Brüggemann | EVELIQ Tracevor 9 Tagen

Interesting demonstration of task-specific decision models! Microsoft-Decision-1 and OpenAI’s Decisions API show how specialized models can support faster, structured agent decisions. The next step is reproducible benchmarking under comparable conditions. DEMO PERFORMANCE ≠ GENERAL PERFORMANCE VERIFIED Promising progress for efficient AI systems from both teams!

Profilbild von Mr crazy
Mr crazyvor 9 Tagen

@Microsoft Bro Always include Laya And D1 from Liquid Please So we can see complete table

Profilbild von sharon
sharonvor 9 Tagen

@Microsoft This demo is so bad.. with a few simple if statements I can do it endlessly cheaper and faster.. this is not the purpose of a decision model

Ähnliche Videos

Local Liquid d1 3B crushed OpenAI Decision at Snake 🐍 Liquid AI d1 on a single RTX 3090 decides in 8 ms so its snake made 82 moves a second and ate 86 apples while OpenAI's API took 345 ms a call and ate 3 Run decision models ->
0:22

Local Liquid d1 3B crushed OpenAI Decision at Snake 🐍 Liquid AI d1 on a single RTX 3090 decides in 8 ms so its snake made 82 moves a second and ate 86 apples while OpenAI's API took 345 ms a call and ate 3 Run decision models ->

atomic.chat

194,199 Aufrufe • vor 12 Tagen

Local Laya beat OpenAI's decision API 🕷 Four decision models raced down a web page ripping out links that name a spider and Laya running locally on an M5 Max took 34 ms a link so it checked 3x more than the cloud models Run decision models locally ->
0:30

Local Laya beat OpenAI's decision API 🕷 Four decision models raced down a web page ripping out links that name a spider and Laya running locally on an M5 Max took 34 ms a link so it checked 3x more than the cloud models Run decision models locally ->

atomic.chat

216,880 Aufrufe • vor 12 Tagen

Next up on First Pass: a conversation with Sam Crowder from LangChain on decision models. Jev from TypeSafe AI took the world by storm. Then OpenAI and Databricks released producs (Decisions API and ai_decide function) So what are Decision Models? We dig in!
11:11

Next up on First Pass: a conversation with Sam Crowder from LangChain on decision models. Jev from TypeSafe AI took the world by storm. Then OpenAI and Databricks released producs (Decisions API and ai_decide function) So what are Decision Models? We dig in!

Jamin Ball

12,147 Aufrufe • vor 11 Tagen

Local Laya blitzed other models on RTX 4090 ⚡️ Four local decision models raced down Wikipedia picking out every word that names a centipede and Laya took 3.9 ms a word so it ended up being the fastest overall Run decision models locally ->
0:20

Local Laya blitzed other models on RTX 4090 ⚡️ Four local decision models raced down Wikipedia picking out every word that names a centipede and Laya took 3.9 ms a word so it ended up being the fastest overall Run decision models locally ->

atomic.chat

26,773 Aufrufe • vor 11 Tagen

built a decision journal AI app inspired by Farnam Street 's decision journal concept over the weekend. it runs locally on your laptop, is encrypted and it helps you make high-consequence low-conviction decisions faster and learn from them methodically
0:36

built a decision journal AI app inspired by Farnam Street 's decision journal concept over the weekend. it runs locally on your laptop, is encrypted and it helps you make high-consequence low-conviction decisions faster and learn from them methodically

Sina Meraji

16,197 Aufrufe • vor 6 Monaten

Introducing DecisionModels Decision Models I've spent the last four weeks benchmarking decision models (I run JevBench). Today I'm launching the place to actually use the best ones: ➡️Top open models ➡️Up to 4x faster on our own REAX engine ➡️From $0.019 per 1M input tokens. Expect more reveals soon, and read 👇
0:30

Introducing DecisionModels Decision Models I've spent the last four weeks benchmarking decision models (I run JevBench). Today I'm launching the place to actually use the best ones: ➡️Top open models ➡️Up to 4x faster on our own REAX engine ➡️From $0.019 per 1M input tokens. Expect more reveals soon, and read 👇

Florian S

1,095,711 Aufrufe • vor 5 Tagen

One of Floyd Mayweather’s most dominant performances yet somehow it was scored a majority decision Canelo was 23 years old, undefeated at 42-0-1 and as hungry as ever to be the next superstar of boxing.. I still can’t get over the fact one judge scored this fight 114-114 🤯
2:35

One of Floyd Mayweather’s most dominant performances yet somehow it was scored a majority decision Canelo was 23 years old, undefeated at 42-0-1 and as hungry as ever to be the next superstar of boxing.. I still can’t get over the fact one judge scored this fight 114-114 🤯

BoxingEnthusiast

213,884 Aufrufe • vor 1 Jahr

1. The Pardon Board's decision to reduce Najib Razak's jail sentence and fine is due to the 'mercy and compassion' of the 16th Yang di-Pertuan Agong, says Anwar Ibrahim. "The decision by the (former) King is final and it is based on significant considerations, not just the law."
7:21

1. The Pardon Board's decision to reduce Najib Razak's jail sentence and fine is due to the 'mercy and compassion' of the 16th Yang di-Pertuan Agong, says Anwar Ibrahim. "The decision by the (former) King is final and it is based on significant considerations, not just the law."

BFM News

548,157 Aufrufe • vor 2 Jahren

We just launched Telnyx Decision Models in beta. AI agents make small judgment calls on every run. Is this caller asking for a refund? Which team owns this ticket? Is this incident urgent enough to page someone? Decision Models answers them in one API call. Send the context and up to 64 typed questions, and get typed answers back: a choice, a yes/no score, or a rating on a scale. Your code branches on the answer. There's no free text to parse. We are offering 2 models, chosen per request: → Decision Flash, the default, for high-volume routing and triage at the lowest cost and latency. $0.035 per 1M input tokens. → Decision Pro, for decisions that depend on long context, like a full support history or a long call transcript. $0.20 per 1M input tokens. Output tokens are free on both. The API is TypeSafe-compatible, so teams already using the TypeSafe SDK can point it at Telnyx with a new base URL and model name. It runs on the same API key as Telnyx voice, speech-to-text, and inference, so you don’t need to join any waitlist or go through extra setup to access it. Try it today:
0:39

We just launched Telnyx Decision Models in beta. AI agents make small judgment calls on every run. Is this caller asking for a refund? Which team owns this ticket? Is this incident urgent enough to page someone? Decision Models answers them in one API call. Send the context and up to 64 typed questions, and get typed answers back: a choice, a yes/no score, or a rating on a scale. Your code branches on the answer. There's no free text to parse. We are offering 2 models, chosen per request: → Decision Flash, the default, for high-volume routing and triage at the lowest cost and latency. $0.035 per 1M input tokens. → Decision Pro, for decisions that depend on long context, like a full support history or a long call transcript. $0.20 per 1M input tokens. Output tokens are free on both. The API is TypeSafe-compatible, so teams already using the TypeSafe SDK can point it at Telnyx with a new base URL and model name. It runs on the same API key as Telnyx voice, speech-to-text, and inference, so you don’t need to join any waitlist or go through extra setup to access it. Try it today:

Telnyx

866,183 Aufrufe • vor 25 Tagen

I’ve been using AI to help my decision-making for over 30 years. It’s a big part of what made Bridgewater successful. I found that if I wrote down my criteria for making decisions and put them into algorithms, the computer could process far more information far more quickly than I could on my own. Alongside my logic and inspiration, it was a great partnership. I started doing this in my investment decision-making, but soon incorporated it into my broader decision-making, as well. (By the way, that investment decision-making is what made Bridgewater so successful.) LLMs have made this process even more seamless and useful.
1:28

I’ve been using AI to help my decision-making for over 30 years. It’s a big part of what made Bridgewater successful. I found that if I wrote down my criteria for making decisions and put them into algorithms, the computer could process far more information far more quickly than I could on my own. Alongside my logic and inspiration, it was a great partnership. I started doing this in my investment decision-making, but soon incorporated it into my broader decision-making, as well. (By the way, that investment decision-making is what made Bridgewater so successful.) LLMs have made this process even more seamless and useful.

Ray Dalio

118,235 Aufrufe • vor 10 Monaten

Undefeated no more ‼️ No. 4 at 157 lbs PJ Duke delivered a 20-7 major decision over previously undefeated No. 1 Brandon Cannon 👀 Watch the #B1GWrestling quadbox only on 💻
0:57

Undefeated no more ‼️ No. 4 at 157 lbs PJ Duke delivered a 20-7 major decision over previously undefeated No. 1 Brandon Cannon 👀 Watch the #B1GWrestling quadbox only on 💻

Big Ten Wrestling

23,393 Aufrufe • vor 7 Monaten

Draymond Green says he knew Philly was LeBrons choice earlier on: “I wasn’t blindsided, i know a few people that know a few people. So it wasn’t surprising. I didn’t know when the announcement would come though. One of their biggest mistakes was ‘The Decision’. I think the decision raised like $4 million for the boys and girls club. And they got crushed for it, but it made a lot of money for charity. And now LeBron just released a video with a caption. There was so much buildup and then it’s just an instagram post. After the first decision yall have Bron scarred man. Because yall crushed him.” - The Draymond Green Show Draymond Green
2:16

Draymond Green says he knew Philly was LeBrons choice earlier on: “I wasn’t blindsided, i know a few people that know a few people. So it wasn’t surprising. I didn’t know when the announcement would come though. One of their biggest mistakes was ‘The Decision’. I think the decision raised like $4 million for the boys and girls club. And they got crushed for it, but it made a lot of money for charity. And now LeBron just released a video with a caption. There was so much buildup and then it’s just an instagram post. After the first decision yall have Bron scarred man. Because yall crushed him.” - The Draymond Green Show Draymond Green

JayTakes

146,498 Aufrufe • vor 2 Monaten

A 9B model from Satya Nadella's Microsoft just beat Sam Altman's GPT-6 Sol at a Rubik's Cube, and it's built on Alibaba's Qwen Same scramble, same 12-move solution for all 4 models, 1 model call per move, real time Results: Microsoft Decision-1: solved in 5.9s for $151 per 1M solves TypeSafe Jev: solved in 6.8s for the same $151 Alibaba Qwen3.8 27B: solved in 10.4s for $5,040, open weights OpenAI GPT-6 Sol: stuck on move 8 when the clip ends, $31,200 per 1M solves OpenAI's flagship burns 206x more money than Microsoft's 9B model to make the same moves, and a free-to-download Qwen running at $0.15 per million tokens still finishes 4x faster than GPT-6 Sol
0:30

A 9B model from Satya Nadella's Microsoft just beat Sam Altman's GPT-6 Sol at a Rubik's Cube, and it's built on Alibaba's Qwen Same scramble, same 12-move solution for all 4 models, 1 model call per move, real time Results: Microsoft Decision-1: solved in 5.9s for $151 per 1M solves TypeSafe Jev: solved in 6.8s for the same $151 Alibaba Qwen3.8 27B: solved in 10.4s for $5,040, open weights OpenAI GPT-6 Sol: stuck on move 8 when the clip ends, $31,200 per 1M solves OpenAI's flagship burns 206x more money than Microsoft's 9B model to make the same moves, and a free-to-download Qwen running at $0.15 per million tokens still finishes 4x faster than GPT-6 Sol

leopardracer

22,981 Aufrufe • vor 8 Tagen

Laya plays Plants vs Zombies on CPU 🧟 We patched the llama.cpp engine to run Laya natively so it works on a regular CPU with no GPU and here a fine-tuned Laya makes every move until it beats the Plants vs Zombies pool level Run decision models locally->
0:55

Laya plays Plants vs Zombies on CPU 🧟 We patched the llama.cpp engine to run Laya natively so it works on a regular CPU with no GPU and here a fine-tuned Laya makes every move until it beats the Plants vs Zombies pool level Run decision models locally->

atomic.chat

14,071 Aufrufe • vor 17 Tagen

Was Keir Starmer responsible for the decision NOT to prosecute Jimmy Saville? Keir Starmer was head of the CPS when the decision was taken not to prosecute Saville in 2009 but the decision was taken by a local CPS lawyer who decided there was not enough evidence to proceed. So he maintained it was not his decision. He later apologised after a review in 2011 (once Saville died & it all came out) for the CPS shortcomings but maintained he was not personally involved. At the end of the day the buck stops at the top & HE WAS THE HEAD OF THE CPS, so he has to take responsibility. In the video below he says: “I take full responsibility for every decision of the Crown Prosecution Service while I was Director of Public Prosecutions.” There you have it straight from the horse’s mouth. YES he was responsible for the failure to prosecute Britains most notorious & prolific child sex offender.
0:11

Was Keir Starmer responsible for the decision NOT to prosecute Jimmy Saville? Keir Starmer was head of the CPS when the decision was taken not to prosecute Saville in 2009 but the decision was taken by a local CPS lawyer who decided there was not enough evidence to proceed. So he maintained it was not his decision. He later apologised after a review in 2011 (once Saville died & it all came out) for the CPS shortcomings but maintained he was not personally involved. At the end of the day the buck stops at the top & HE WAS THE HEAD OF THE CPS, so he has to take responsibility. In the video below he says: “I take full responsibility for every decision of the Crown Prosecution Service while I was Director of Public Prosecutions.” There you have it straight from the horse’s mouth. YES he was responsible for the failure to prosecute Britains most notorious & prolific child sex offender.

J Stewart

270,426 Aufrufe • vor 1 Jahr

lando: “it obviously was not a bad decision for lap 1, it worked out in a way. at the same time it was the wrong decision. so we have to accept that.. but i also pushed for it myself quite a bit. so we take it on the chin” “but we had some problems and we had a failure in the end which put us out. so yeah, put us out of misery in a way” 🙃🙃
0:46

lando: “it obviously was not a bad decision for lap 1, it worked out in a way. at the same time it was the wrong decision. so we have to accept that.. but i also pushed for it myself quite a bit. so we take it on the chin” “but we had some problems and we had a failure in the end which put us out. so yeah, put us out of misery in a way” 🙃🙃

ray

209,687 Aufrufe • vor 4 Monaten

Odablock Warriors go on a legendary run of uniques at Moons and Barrows after making the tough decision to skip the breach on day 2 of DMM All-Stars and it paid off We are so back
0:54

Odablock Warriors go on a legendary run of uniques at Moons and Barrows after making the tough decision to skip the breach on day 2 of DMM All-Stars and it paid off We are so back

Odablock Clips

30,219 Aufrufe • vor 4 Monaten