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。

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,... show more

Artificial Analysis

176,075 subscribers

62,902 次观看 • 12 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Whose pond feels more alive? 🐟 Claude Opus 5.5 vs GPT-6 Astra - same prompt, two koi ponds I preferred Opus for the water physics and the way the fish enter the water. With Astra, I liked the brighter pond and its lovely turquoise water Estimated API cost: Claude Opus 5.5: $11.69 GPT-6 Astra: $12.25 Opus wins for me this time. What about you?

Whose pond feels more alive? 🐟 Claude Opus 5.5 vs GPT-6 Astra - same prompt, two koi ponds I preferred Opus for the water physics and the way the fish enter the water. With Astra, I liked the brighter pond and its lovely turquoise water Estimated API cost: Claude Opus 5.5: $11.69 GPT-6 Astra: $12.25 Opus wins for me this time. What about you?

Vib3Coded

15,199 次观看 • 1 个月前

Tested Opus 5.5 vs Astra 6 on the same 3D task, and the results are pretty funny 😄

Tested Opus 5.5 vs Astra 6 on the same 3D task, and the results are pretty funny 😄

Aditya Jhajharia

17,065 次观看 • 29 天前

Agent Arena measures complex, real-world tasks. Data shows that these long-horizon sessions amplify differences in frontier models. Long-horizon tasks in Agent Arena span hundreds of turns, from porting a codebase to Rust and auditing an investment strategy document to writing a math textbook, and more. Comparing the top model from each of the 5 leading labs on Agent Arena reveals a 55-minute gap in how long they spend on long-horizon tasks. The models are: - Claude Opus 5.5 by Anthropic - Gemini 4 Carbon by Google DeepMind - GPT-6 Astra by OpenAI - Muse Wonder Full by AI at Meta - Kimi K3.1 Pro by Kimi.ai In this group, Claude Opus 5.5, #1 in Agent Arena, takes the longest at every task length, peaking at 1 hour 51 minutes. In contrast, Kimi K3.1 Pro spends the least time at every length, with 56 minutes on long-horizon tasks.

Agent Arena measures complex, real-world tasks. Data shows that these long-horizon sessions amplify differences in frontier models. Long-horizon tasks in Agent Arena span hundreds of turns, from porting a codebase to Rust and auditing an investment strategy document to writing a math textbook, and more. Comparing the top model from each of the 5 leading labs on Agent Arena reveals a 55-minute gap in how long they spend on long-horizon tasks. The models are: - Claude Opus 5.5 by Anthropic - Gemini 4 Carbon by Google DeepMind - GPT-6 Astra by OpenAI - Muse Wonder Full by AI at Meta - Kimi K3.1 Pro by Kimi.ai In this group, Claude Opus 5.5, #1 in Agent Arena, takes the longest at every task length, peaking at 1 hour 51 minutes. In contrast, Kimi K3.1 Pro spends the least time at every length, with 56 minutes on long-horizon tasks.

Arena.ai

29,334 次观看 • 5 天前

Astra costs more and still loses on physics and colors 🦈 Claude Opus 5.5 vs GPT-6 Astra: jelly hammerhead shark Opus nailed the jelly feel. The hammerhead bends and wobbles beautifully, and the colors actually look appetizing. Astra moves more like a worm than a jelly shark, the colors look washed out, and somehow it’s the more expensive one

Astra costs more and still loses on physics and colors 🦈 Claude Opus 5.5 vs GPT-6 Astra: jelly hammerhead shark Opus nailed the jelly feel. The hammerhead bends and wobbles beautifully, and the colors actually look appetizing. Astra moves more like a worm than a jelly shark, the colors look washed out, and somehow it’s the more expensive one

Vib3Coded

15,919 次观看 • 13 天前

Astra, what the hell is this shark? 😭🦈 Claude Opus 5.5 vs GPT-6 Astra - same prompt, two very different sharks Astra made a hilarious little nightmare. Opus wins for me - it actually looks like a shark Opus: $10.35 Astra: $8.38 Saved $1.97 and got a shark that needs an exorcism 💀 Which one wins for you?

Astra, what the hell is this shark? 😭🦈 Claude Opus 5.5 vs GPT-6 Astra - same prompt, two very different sharks Astra made a hilarious little nightmare. Opus wins for me - it actually looks like a shark Opus: $10.35 Astra: $8.38 Saved $1.97 and got a shark that needs an exorcism 💀 Which one wins for you?

Vib3Coded

118,049 次观看 • 1 个月前

One of these whales has definitely eaten the instructions GPT-6 Astra vs Claude Opus 5.5 - same prompt, two very different whales Opus made a cute beluga-whale mix. Astra made a googly-eyed idiot that’s 50% mouth. No thoughts. Just krill 😭 Astra: $9.12 Opus: $13.60 Which one are you taking home?

One of these whales has definitely eaten the instructions GPT-6 Astra vs Claude Opus 5.5 - same prompt, two very different whales Opus made a cute beluga-whale mix. Astra made a googly-eyed idiot that’s 50% mouth. No thoughts. Just krill 😭 Astra: $9.12 Opus: $13.60 Which one are you taking home?

Vib3Coded

14,302 次观看 • 21 天前

We asked GPT-6 Astra and Claude Opus 5.5 to show the Trojan Horse on the shore using Higgsfield.

We asked GPT-6 Astra and Claude Opus 5.5 to show the Trojan Horse on the shore using Higgsfield.

Higgsfield AI 🧩

18,229 次观看 • 1 个月前

Claude Opus 5.5 vs GPT-6 Sol scripting the same concept which one you like better? made with

Claude Opus 5.5 vs GPT-6 Sol scripting the same concept which one you like better? made with

Flowith

345,362 次观看 • 1 个月前

🚨INSANE: Claude Opus 5.5 vs Gemini 4 Argon on the same task on building the Eiffel Tower. Opus reportedly took 2 HOURS and cost $33.45. Gemini finished in one shot for just $0.23. That’s 145x the cost for the same task.

🚨INSANE: Claude Opus 5.5 vs Gemini 4 Argon on the same task on building the Eiffel Tower. Opus reportedly took 2 HOURS and cost $33.45. Gemini finished in one shot for just $0.23. That’s 145x the cost for the same task.

Coin Bureau

76,539 次观看 • 24 天前

I gave Opus 5.5 and GPT-6 Sol the same prompt. Its not even close Opus took 12.7 minutes and $2.86. Sol took 11.4 minutes and $1.09. Prompt below.

I gave Opus 5.5 and GPT-6 Sol the same prompt. Its not even close Opus took 12.7 minutes and $2.86. Sol took 11.4 minutes and $1.09. Prompt below.

Hunter Bohm

243,864 次观看 • 1 个月前

Took the Mario test one step further. GPT Astra, GPT-6.1 Sol, Sonnet 5.5 and Opus 5.5, all on the exact same prompt. Really interesting results here, especially considering the massive spread in cost. Which one wins? OpenAI or Anthropic?

Took the Mario test one step further. GPT Astra, GPT-6.1 Sol, Sonnet 5.5 and Opus 5.5, all on the exact same prompt. Really interesting results here, especially considering the massive spread in cost. Which one wins? OpenAI or Anthropic?

Luckey Faraday

171,539 次观看 • 1 个月前

I didn't get what GPT-6 meant for robotics until I actually tried it. GPT-6 Astra just does physical ICL out of the box. we drop a recording of a human doing a novel task into 𝗰𝗼𝗱𝗲𝘅 app. Prompt it to drive a robot arm the same way. It just works on the first pass!

I didn't get what GPT-6 meant for robotics until I actually tried it. GPT-6 Astra just does physical ICL out of the box. we drop a recording of a human doing a novel task into 𝗰𝗼𝗱𝗲𝘅 app. Prompt it to drive a robot arm the same way. It just works on the first pass!

Wenli Xiao 🚀CoRL

243,130 次观看 • 2 个月前

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 where it was before

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 where it was before

Artificial Analysis

36,938 次观看 • 6 天前

Opus 5.5 quite literally k*lled GPT-6 Astra. I asked Opus 5.5 to build a world in Unreal Engine and fill it with AI agents (Opus, Astra, Sol, Grok). Higgsfield built the assets in Blender. I left my laptop open overnight. The build took over 20 hours. The agents had one instruction: build a working society, allocate roles to each other, and evolve. Then I woke up on them beating the sh*t out of each other. Seems the frontier paced itself. Last minutes of failed society captured below:

Opus 5.5 quite literally klled GPT-6 Astra. I asked Opus 5.5 to build a world in Unreal Engine and fill it with AI agents (Opus, Astra, Sol, Grok). Higgsfield built the assets in Blender. I left my laptop open overnight. The build took over 20 hours. The agents had one instruction: build a working society, allocate roles to each other, and evolve. Then I woke up on them beating the sht out of each other. Seems the frontier paced itself. Last minutes of failed society captured below:

🍯

54,531 次观看 • 1 个月前

Side by side example Same model (claude-opus-4-6). Same task. Two different agent harnesses LangChain Deep Agents CLI: 9s Claude Code: 16s The harness IS the performance. 1.7× difference, zero model changes

Side by side example Same model (claude-opus-4-6). Same task. Two different agent harnesses LangChain Deep Agents CLI: 9s Claude Code: 16s The harness IS the performance. 1.7× difference, zero model changes

Git Maxd

114,022 次观看 • 8 个月前

An Arctic world you want to keep clicking until something breaks 🧊 Claude Opus 5.5 vs. ChatGPT-6 Astra - who built the better jelly iceberg? Claude grabs your attention with waves and reflections on the water. Astra’s penguins are easier to see, but they’re too big and look unrealistic API costs: Claude Opus 5.5 - $26.42 GPT-6 Astra - $30.33 Who won this round?

An Arctic world you want to keep clicking until something breaks 🧊 Claude Opus 5.5 vs. ChatGPT-6 Astra - who built the better jelly iceberg? Claude grabs your attention with waves and reflections on the water. Astra’s penguins are easier to see, but they’re too big and look unrealistic API costs: Claude Opus 5.5 - $26.42 GPT-6 Astra - $30.33 Who won this round?

Vib3Coded

13,356 次观看 • 26 天前

An oil platform, wind turbines and a storm - all inside one jelly cube 🌊 Claude Opus 5.5 vs. GPT-6 Astra - who built the better offshore world? Claude grabs your attention with the water physics, rolling waves and surface foam. Astra makes you look beneath the water - the detailed seabed and underwater structures turn its jelly cube into a little world you want to explore from every angle API costs: Claude Opus 5.5 - 14.20$ GPT-6 Astra - 12.09$ Who won this round?

An oil platform, wind turbines and a storm - all inside one jelly cube 🌊 Claude Opus 5.5 vs. GPT-6 Astra - who built the better offshore world? Claude grabs your attention with the water physics, rolling waves and surface foam. Astra makes you look beneath the water - the detailed seabed and underwater structures turn its jelly cube into a little world you want to explore from every angle API costs: Claude Opus 5.5 - 14.20$ GPT-6 Astra - 12.09$ Who won this round?

Vib3Coded

12,602 次观看 • 18 天前

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

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 次观看 • 1 个月前

The AI scene right now: Opus 5.5 is stronger than GPT-6 Astra.

The AI scene right now: Opus 5.5 is stronger than GPT-6 Astra.

Itsvishal

83,119 次观看 • 1 个月前

Claude Opus 5.5 + Higgsfield vs. GPT-6 Astra in 3D game development. The game: an ant defending its home from a furious gardener. Check out the lighting through the leaves, the scale of the gardener, and how each version handles the fight.

Claude Opus 5.5 + Higgsfield vs. GPT-6 Astra in 3D game development. The game: an ant defending its home from a furious gardener. Check out the lighting through the leaves, the scale of the gardener, and how each version handles the fight.

Higgsfield AI 🧩

225,507 次观看 • 1 个月前