Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Knowing when to stop is hard for language models controlling a robot arm ➤ Claude Opus 5.5 completed the task in 23 minutes, and said so: "The crocodile looks fully inside now" ➤ It then spent 22 minutes pushing the crocodile back and forth, only to finish almost exactly...

ArtificialAnlys's profile picture

Artificial Analysis

175,902 subscribers

36,938 Aufrufe • vor 6 Tagen •via X (Twitter)

Wissenschaft & Technologie
LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

When you listen to what Chris Hani believed in, then compare it to the things Nelson Mandela and co were pushing for, you can clearly see the crocodile tears when Hani was murdered
0:15

When you listen to what Chris Hani believed in, then compare it to the things Nelson Mandela and co were pushing for, you can clearly see the crocodile tears when Hani was murdered

SizweLo

29,880 Aufrufe • vor 22 Tagen

Antigravity is now quite insane You can use Claude Opus 5.5 and Sonnet 5.5 directly inside it... in addition to Gemini 3.8 Flash. So basically my workflow is now: - Opus 5.5 for planning - Sonnet 5.5 for frontend - Gemini 3.8 Flash to execute Everything in the same place and without paying an additional subscription.
0:18

Antigravity is now quite insane You can use Claude Opus 5.5 and Sonnet 5.5 directly inside it... in addition to Gemini 3.8 Flash. So basically my workflow is now: - Opus 5.5 for planning - Sonnet 5.5 for frontend - Gemini 3.8 Flash to execute Everything in the same place and without paying an additional subscription.

Paul Couvert

47,987 Aufrufe • vor 1 Monat

This took me 10 minutes with Opus 5.5 It's f*cking unreal… Upload the PDF below to Opus 5.5 and try it yourself. Then read the article on Opus 5.5 best practices for motion design.
0:14

This took me 10 minutes with Opus 5.5 It's f*cking unreal… Upload the PDF below to Opus 5.5 and try it yourself. Then read the article on Opus 5.5 best practices for motion design.

nicco

57,177 Aufrufe • vor 11 Tagen

After Andrej Karpathy released autoresearch, where an agent runs experiments in a loop on its own, I packed his method into one prompt that turns Opus 5.5 into a graph and can carry almost any task to the finish. → writes a check that can actually fail, first → splits the task into parts → runs the independent parts in parallel → loops each one until it passes the check → sends back only the part that failed → turns every mistake into a rule for next time → calls you only for the step that can’t be undone Throw it at Opus 5.5 with the task you’ve been putting off for a week, and you’ll see the difference in one run. The full prompt is in the comments. How to tune it for your own work is in the article below. Save it before it gets lost ↓
0:15

After Andrej Karpathy released autoresearch, where an agent runs experiments in a loop on its own, I packed his method into one prompt that turns Opus 5.5 into a graph and can carry almost any task to the finish. → writes a check that can actually fail, first → splits the task into parts → runs the independent parts in parallel → loops each one until it passes the check → sends back only the part that failed → turns every mistake into a rule for next time → calls you only for the step that can’t be undone Throw it at Opus 5.5 with the task you’ve been putting off for a week, and you’ll see the difference in one run. The full prompt is in the comments. How to tune it for your own work is in the article below. Save it before it gets lost ↓

Hoodie

326,615 Aufrufe • vor 5 Tagen

Same robot, same task, two models with very different writing styles ➤ On the left, Claude Opus 5.5 dabs out "hello world" dot by dot: 3,270 arm moves over 34 minutes ➤ On the right, GPT-6 Astra writes in continuous strokes: 514 moves in 13 minutes From AA-Robotics v0.1, seen from the model’s view See more examples here:
0:11

Same robot, same task, two models with very different writing styles ➤ On the left, Claude Opus 5.5 dabs out "hello world" dot by dot: 3,270 arm moves over 34 minutes ➤ On the right, GPT-6 Astra writes in continuous strokes: 514 moves in 13 minutes From AA-Robotics v0.1, seen from the model’s view See more examples here:

Artificial Analysis

62,902 Aufrufe • vor 12 Tagen

THIS IS F*CKING BRILLIANT. Anthropic’s docs describe one of the smartest ways to run Opus three models work on one task and each does only the part it’s cheapest at. start Claude Code like this: claude -> model sonnet -> advisor opus you stop paying Opus prices for work that doesn’t need Opus > Sonnet runs the session and edits the code > Haiku 5.5 searches the repo and docs, at up to 40x less per token than Opus 5.5 > Opus waits as the advisor and reads the whole session, tool calls included Claude usually calls Opus at three points: > before it locks a plan > when the same error comes back > right before it says it’s done Sonnet drives. Haiku scouts. Opus signs off and if you go one step lower, most of an agent’s day is tiny choices like picking a file or a tool, or deciding whether to retry. Jev takes those first, so Sonnet and Opus only see the moments that are really open the whole setup Sonnet runs the main session Explore on Haiku 5.5 searches and pulls docs, read-only Opus on call as the advisor give Claude Code this setup and the prompt below ↓ “Set up my Claude Code exactly like this: Main model Sonnet, advisor Opus, saved as advisorModel in my settings. One read-only subagent named exactly Explore, on claude-haiku-5-5, to replace the built-in one. One line in CLAUDE.md: ask the advisor before locking a plan, after the same error twice, and before calling a task done, not more often. Check what could turn the advisor off (CLAUDE_CODE_DISABLE_ADVISOR_TOOL, DISABLE_TELEMETRY) and only report it. Show every change as a diff first. Don’t edit anything until I say go.” before you turn it on: every advisor call re-reads the whole transcript with no cache. on a 200k session that’s about $0.80 in input, against $0.04 for a normal request. keep it to a few calls the built-in Explore agent runs on your main model unless a subagent is named exactly Explore. any other name does nothing turning the advisor on or off keeps your prompt cache. switching models rebuilds it it needs the Anthropic API, and it stays off if telemetry is disabled bookmark it before your next long session
0:17

THIS IS F*CKING BRILLIANT. Anthropic’s docs describe one of the smartest ways to run Opus three models work on one task and each does only the part it’s cheapest at. start Claude Code like this: claude -> model sonnet -> advisor opus you stop paying Opus prices for work that doesn’t need Opus > Sonnet runs the session and edits the code > Haiku 5.5 searches the repo and docs, at up to 40x less per token than Opus 5.5 > Opus waits as the advisor and reads the whole session, tool calls included Claude usually calls Opus at three points: > before it locks a plan > when the same error comes back > right before it says it’s done Sonnet drives. Haiku scouts. Opus signs off and if you go one step lower, most of an agent’s day is tiny choices like picking a file or a tool, or deciding whether to retry. Jev takes those first, so Sonnet and Opus only see the moments that are really open the whole setup Sonnet runs the main session Explore on Haiku 5.5 searches and pulls docs, read-only Opus on call as the advisor give Claude Code this setup and the prompt below ↓ “Set up my Claude Code exactly like this: Main model Sonnet, advisor Opus, saved as advisorModel in my settings. One read-only subagent named exactly Explore, on claude-haiku-5-5, to replace the built-in one. One line in CLAUDE.md: ask the advisor before locking a plan, after the same error twice, and before calling a task done, not more often. Check what could turn the advisor off (CLAUDE_CODE_DISABLE_ADVISOR_TOOL, DISABLE_TELEMETRY) and only report it. Show every change as a diff first. Don’t edit anything until I say go.” before you turn it on: every advisor call re-reads the whole transcript with no cache. on a 200k session that’s about $0.80 in input, against $0.04 for a normal request. keep it to a few calls the built-in Explore agent runs on your main model unless a subagent is named exactly Explore. any other name does nothing turning the advisor on or off keeps your prompt cache. switching models rebuilds it it needs the Anthropic API, and it stays off if telemetry is disabled bookmark it before your next long session

rvaniaaa

62,223 Aufrufe • vor 27 Tagen

🚨 THIS IS HOW IT STARTS — HUMANOID ROBOT SUDDENLY LUNGES AT A MAN — EXACTLY THE MOMENT PEOPLE FEARED A humanoid robot was casually kicking a soccer ball around during a public demonstration. Then the ball rolled toward a man in the audience. He tried to kick it back. That’s when the robot suddenly charged toward him and raised its arm like it was about to swing. Within seconds, a handler rushed in and grabbed the robot before the moment escalated. People in the room laugh nervously… but the clip is making the rounds online for one reason: The movement looks almost human. The way it runs. The stance. The arm coming up. Even the comments are losing it: • “Tf was it going to HIT for???” • “Why did it need to be held back?” • “DON’T TOUCH MY BALL” • “How do they get pissed off if they supposedly don’t have emotions?” The strangest part? The robot seemed to react the instant the human interfered with the ball. Are we just watching a glitch during a demo… or the exact moment people warned about?
0:16

🚨 THIS IS HOW IT STARTS — HUMANOID ROBOT SUDDENLY LUNGES AT A MAN — EXACTLY THE MOMENT PEOPLE FEARED A humanoid robot was casually kicking a soccer ball around during a public demonstration. Then the ball rolled toward a man in the audience. He tried to kick it back. That’s when the robot suddenly charged toward him and raised its arm like it was about to swing. Within seconds, a handler rushed in and grabbed the robot before the moment escalated. People in the room laugh nervously… but the clip is making the rounds online for one reason: The movement looks almost human. The way it runs. The stance. The arm coming up. Even the comments are losing it: • “Tf was it going to HIT for???” • “Why did it need to be held back?” • “DON’T TOUCH MY BALL” • “How do they get pissed off if they supposedly don’t have emotions?” The strangest part? The robot seemed to react the instant the human interfered with the ball. Are we just watching a glitch during a demo… or the exact moment people warned about?

HustleBitch

54,191 Aufrufe • vor 7 Monaten

ANTHROPIC JUST DROPPED THE OFFICIAL GUIDE TO PROMPTING OPUS 5 This is the most important prompting guide for anyone using Claude. Bookmark this before you forget. Most people still prompt Opus 5 like older models. That's the mistake. > give the complete task upfront — don't drip-feed instructions. > explicitly ask for shorter responses if you want them. > use low/medium effort by default. increase effort only when needed. > let it finish the entire feature instead of asking for partial implementations. > use subagents intentionally for large codebases, not every task. Opus 5 is at its best when you stop micromanaging it and let it own the work. Read it before writing your next prompt. Claude → Opus 5 → Agentic Execution → Production-Ready Output → Money
0:11

ANTHROPIC JUST DROPPED THE OFFICIAL GUIDE TO PROMPTING OPUS 5 This is the most important prompting guide for anyone using Claude. Bookmark this before you forget. Most people still prompt Opus 5 like older models. That's the mistake. > give the complete task upfront — don't drip-feed instructions. > explicitly ask for shorter responses if you want them. > use low/medium effort by default. increase effort only when needed. > let it finish the entire feature instead of asking for partial implementations. > use subagents intentionally for large codebases, not every task. Opus 5 is at its best when you stop micromanaging it and let it own the work. Read it before writing your next prompt. Claude → Opus 5 → Agentic Execution → Production-Ready Output → Money

Rahul

408,531 Aufrufe • vor 3 Monaten

The scariest thing about Opus 5.5 is that it goes back to fill in the arm realising the lines were dotted.
0:15

The scariest thing about Opus 5.5 is that it goes back to fill in the arm realising the lines were dotted.

H

418,384 Aufrufe • vor 1 Monat

Motion design in 2026: > Open Claude Opus 5.5. > Launch After Effects. > Let Claude run the scripts through Higgsfield. > Come back in 30 minutes. Then review the render, adjust the pacing, make fixes and give it another pass. Repeat until you’re satisfied.
0:12

Motion design in 2026: > Open Claude Opus 5.5. > Launch After Effects. > Let Claude run the scripts through Higgsfield. > Come back in 30 minutes. Then review the render, adjust the pacing, make fixes and give it another pass. Repeat until you’re satisfied.

Higgsfield AI 🧩

24,919 Aufrufe • vor 1 Monat

asked Claude Opus 5.5 and GPT-6 Sol to create the glitter sticker effect seems like Opus 5.5 ate this task
0:12

asked Claude Opus 5.5 and GPT-6 Sol to create the glitter sticker effect seems like Opus 5.5 ate this task

Ann Nguyen

472,925 Aufrufe • vor 1 Monat

At the zoo, everyone was left terrified when a crocodile suddenly attacked a helpless child near the water. As the crocodile moved closer, the child cried out for help and people could only watch in fear. Just when the situation seemed impossible to control, a brave elephant rushed forward and confronted the crocodile. The elephant fearlessly stood between the child and the predator, forcing the crocodile to move away. In that critical moment, the elephant became the child’s protector and saved his life. 🐘❤️
0:15

At the zoo, everyone was left terrified when a crocodile suddenly attacked a helpless child near the water. As the crocodile moved closer, the child cried out for help and people could only watch in fear. Just when the situation seemed impossible to control, a brave elephant rushed forward and confronted the crocodile. The elephant fearlessly stood between the child and the predator, forcing the crocodile to move away. In that critical moment, the elephant became the child’s protector and saved his life. 🐘❤️

Ajit sah

51,195 Aufrufe • vor 2 Monaten

Opus 5.5 is a motion designer now It’s f*cking unreal… Give the PDF below to Opus 5.5 and watch it work. Then read the guide below, it’s the easiest way to get started.
0:14

Opus 5.5 is a motion designer now It’s f*cking unreal… Give the PDF below to Opus 5.5 and watch it work. Then read the guide below, it’s the easiest way to get started.

seeco

166,466 Aufrufe • vor 13 Tagen

Hayi, FlySafair has been the absolute pits this afternoon from Durban. At first, I was like, “wow, that’s the best landing I’ve had from them in a while.” Then when it was taxiing to the gate and got to the intersection, it braked so hard it was screeching. Ok. Fine. Then we disembark. Is the door not locked and we can’t get inside the airport? Ok. Fine still. Get to the conveyor belt to collect luggage. It works for about 3 minutes. Now it’s been down for 10 minutes and no one is telling us anything. Hayi andina bhongo. No bhongo at all.
0:10

Hayi, FlySafair has been the absolute pits this afternoon from Durban. At first, I was like, “wow, that’s the best landing I’ve had from them in a while.” Then when it was taxiing to the gate and got to the intersection, it braked so hard it was screeching. Ok. Fine. Then we disembark. Is the door not locked and we can’t get inside the airport? Ok. Fine still. Get to the conveyor belt to collect luggage. It works for about 3 minutes. Now it’s been down for 10 minutes and no one is telling us anything. Hayi andina bhongo. No bhongo at all.

Khaya Dlanga

146,615 Aufrufe • vor 1 Monat

Smol tip: you don't have to wait for Claude to finish working before giving it the next task. Type a message and hit Enter to queue it. If you need to edit, ↑ to pull it back into the prompt box.
0:12

Smol tip: you don't have to wait for Claude to finish working before giving it the next task. Type a message and hit Enter to queue it. If you need to edit, ↑ to pull it back into the prompt box.

Delba

44,716 Aufrufe • vor 4 Monaten

Opus 5.5 is the new After Effects and nobody's talking about it… Feed the PDF below to Opus 5.5, you'll thank me later Then start with the guide below, it's all you need for motion design in Opus 5.5
0:12

Opus 5.5 is the new After Effects and nobody's talking about it… Feed the PDF below to Opus 5.5, you'll thank me later Then start with the guide below, it's all you need for motion design in Opus 5.5

nicco

75,119 Aufrufe • vor 19 Tagen

THIS IS PURE F*CKING PLATINUM Anthropic just shared a Claude Code setup that stops you from wasting expensive models on work cheaper ones can handle idea: let Sonnet write the code, Haiku handle the exploration, and Opus review the decisions that actually matter you just start Claude Code like this: `claude --model sonnet --advisor opus` here's how it splits: → Sonnet 5 runs the session and writes the code → Haiku 5.5 explores files and docs in parallel → Opus 5.5 stays on call as advisor for the big decisions → you stop burning Opus context on tasks that don't need it Opus only steps in at three moments: → before a plan locks → when the same error hits twice → before calling it done Sonnet builds. Haiku explores. Opus reviews. JEV engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to JEV in milliseconds, and the big models only see the ones that split full tree > Sonnet 5 runs the main session and edits code > search maps the repo on Haiku 5.5 > grep finds call sites on Haiku 5.5 > docs pulls the API reference on Haiku 5.5 > Opus 5.5 on call as advisor, reads the full transcript when consulted > JEV routes micro-forks in milliseconds paste the tree and this prompt into Claude Code ↓ "Reconfigure my Claude Code setup around this tree: 1. Run the main session on Sonnet 5 with Opus 5.5 as advisor. 2. Create three read-only Haiku 5.5 subagents: search (maps the repo), grep (finds call sites), docs (pulls the API reference). 3. Add one rule to CLAUDE.md: consult the advisor before locking a plan, after the same error twice, and before calling a task done. 4. Route fast deterministic forks (tool choice, path selection, retries) through JEV. Show me every change as a diff first. No edits until I say go." ↳ save it, it'll come in handy
0:15

THIS IS PURE F*CKING PLATINUM Anthropic just shared a Claude Code setup that stops you from wasting expensive models on work cheaper ones can handle idea: let Sonnet write the code, Haiku handle the exploration, and Opus review the decisions that actually matter you just start Claude Code like this: `claude --model sonnet --advisor opus` here's how it splits: → Sonnet 5 runs the session and writes the code → Haiku 5.5 explores files and docs in parallel → Opus 5.5 stays on call as advisor for the big decisions → you stop burning Opus context on tasks that don't need it Opus only steps in at three moments: → before a plan locks → when the same error hits twice → before calling it done Sonnet builds. Haiku explores. Opus reviews. JEV engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to JEV in milliseconds, and the big models only see the ones that split full tree > Sonnet 5 runs the main session and edits code > search maps the repo on Haiku 5.5 > grep finds call sites on Haiku 5.5 > docs pulls the API reference on Haiku 5.5 > Opus 5.5 on call as advisor, reads the full transcript when consulted > JEV routes micro-forks in milliseconds paste the tree and this prompt into Claude Code ↓ "Reconfigure my Claude Code setup around this tree: 1. Run the main session on Sonnet 5 with Opus 5.5 as advisor. 2. Create three read-only Haiku 5.5 subagents: search (maps the repo), grep (finds call sites), docs (pulls the API reference). 3. Add one rule to CLAUDE.md: consult the advisor before locking a plan, after the same error twice, and before calling a task done. 4. Route fast deterministic forks (tool choice, path selection, retries) through JEV. Show me every change as a diff first. No edits until I say go." ↳ save it, it'll come in handy

Livsun

12,338 Aufrufe • vor 28 Tagen

asked opus 5.5 to add animations 5 hours later, here's where it ended up. tried almost everything but it kept breaking the character and animations so i stopped working on it and made cozy lantern hallow instead, which is fully open source now try it now: for zombie game i need help. how should i fix this?
0:18

asked opus 5.5 to add animations 5 hours later, here's where it ended up. tried almost everything but it kept breaking the character and animations so i stopped working on it and made cozy lantern hallow instead, which is fully open source now try it now: for zombie game i need help. how should i fix this?

J A Z I I

13,419 Aufrufe • vor 27 Tagen

He wore a crocodile mask to prank a crocodile, and the croc said 'nice costume, now let's see you run. 😆😆😁
0:14

He wore a crocodile mask to prank a crocodile, and the croc said 'nice costume, now let's see you run. 😆😆😁

Ray Mond

130,230 Aufrufe • vor 1 Monat

Where is the crocodile when we need it 😭😭😭 #videosvuvu
0:11

Where is the crocodile when we need it 😭😭😭 #videosvuvu

Vuvu Videos

30,969 Aufrufe • vor 9 Monaten