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。

Gemini 4 argon vs GPT 6.1 sol both models were tested with same prompt but results came out really different > sol was tested by me in devin cloud at default reasoning > gemini was tested by lentils earlier this month on pre release checkpoint argon isn’t publicly available... show more

J A Z I I

11,060 subscribers

14,442 次观看 • 2 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

Fable 5.1 vs GPT 5.6 Sol tested both models with same prompt at highest reasoning available > Fable took 10 minutes to complete this test and costed $1.6 > Sol took 7 minutes and costed around $.95 even tho sol is cheaper and faster but results are really different which one won here?

Fable 5.1 vs GPT 5.6 Sol tested both models with same prompt at highest reasoning available > Fable took 10 minutes to complete this test and costed $1.6 > Sol took 7 minutes and costed around $.95 even tho sol is cheaper and faster but results are really different which one won here?

J A Z I I

41,089 次观看 • 1 个月前

🚨Gemini 3.6 Flash is trash I tested it on a 3D Golden Gate Bridge, and the results were awful. • I had to re-prompt it three times because it repeatedly ignored the instructions. • First attempt, instead of creating the requested .html file, it first tried to build the experience inside the Gemini app using simulations. • Then second attempt it started placing images from the web into the chat rather than actually producing the file. • Even after getting it to complete the task, the final output was dramatically worse than Gemini 3.1 Pro, which is 5 months old and now not even a top 10 model on leaderboards. This feels like a regression from Gemini 3.5 Flash and honestly, it is one of the weakest models I have tested in the past few months. Has anyone else tested Gemini 3.6 Flash yet, and are you seeing the same thing?

🚨Gemini 3.6 Flash is trash I tested it on a 3D Golden Gate Bridge, and the results were awful. • I had to re-prompt it three times because it repeatedly ignored the instructions. • First attempt, instead of creating the requested .html file, it first tried to build the experience inside the Gemini app using simulations. • Then second attempt it started placing images from the web into the chat rather than actually producing the file. • Even after getting it to complete the task, the final output was dramatically worse than Gemini 3.1 Pro, which is 5 months old and now not even a top 10 model on leaderboards. This feels like a regression from Gemini 3.5 Flash and honestly, it is one of the weakest models I have tested in the past few months. Has anyone else tested Gemini 3.6 Flash yet, and are you seeing the same thing?

Lumina

72,529 次观看 • 2 个月前

GPT-5.5 by Reasoning Effort: I've asked it in Codex to create a physics-based visualisation of RL cycles for different sized models (70b, 1t, 10t), to demonstrate how the amount of RL you can do differs by model size. My assessment of each: - Low: weird slop - Medium: kinda cooked - High: sort of tried but ultimately incoherent - Extra High: elite - really nice idea and well executed Obviously this is just one shot, but worth trying different reasoning levels for the new models, medium seems to be pretty good for GPT-5.5 and it was really bad for many previous GPT models.

GPT-5.5 by Reasoning Effort: I've asked it in Codex to create a physics-based visualisation of RL cycles for different sized models (70b, 1t, 10t), to demonstrate how the amount of RL you can do differs by model size. My assessment of each: - Low: weird slop - Medium: kinda cooked - High: sort of tried but ultimately incoherent - Extra High: elite - really nice idea and well executed Obviously this is just one shot, but worth trying different reasoning levels for the new models, medium seems to be pretty good for GPT-5.5 and it was really bad for many previous GPT models.

Peter Gostev (SF: 22-26 June)

209,258 次观看 • 5 个月前

Gemini Omni doesn’t just allow you to render text more accurately — but to create it in sync with your visuals. 🎥 Choose your type, placement, animation, exposure and more. Prompt for this video: Word by word, one word on the screen at a time: did, you, know, that, this, model, can, do, pretty, good, text!? each word appears with a different animated style, perfect pacing to a rhythm, sizzle reel.

Gemini Omni doesn’t just allow you to render text more accurately — but to create it in sync with your visuals. 🎥 Choose your type, placement, animation, exposure and more. Prompt for this video: Word by word, one word on the screen at a time: did, you, know, that, this, model, can, do, pretty, good, text!? each word appears with a different animated style, perfect pacing to a rhythm, sizzle reel.

Google

153,902 次观看 • 4 个月前

GPT-6 Astra vs Kimi K3. I'm comparing these because the gap between US models and Chinese models is already getting small. If US AI labs slow down, they will be the ones most affected by it. > Astra did it in one shot 6m 20s and $8. > Kimi K3 got almost the same result, but I had to use 2 follow-up prompts. Total: 12 min and $6.35. The results are almost identical. If US labs slow down, China takes the AI race. They will ship Astra-level open-weight models. China is already good at this. Open weights become the default. That's a nightmare for US labs.

GPT-6 Astra vs Kimi K3. I'm comparing these because the gap between US models and Chinese models is already getting small. If US AI labs slow down, they will be the ones most affected by it. > Astra did it in one shot 6m 20s and $8. > Kimi K3 got almost the same result, but I had to use 2 follow-up prompts. Total: 12 min and $6.35. The results are almost identical. If US labs slow down, China takes the AI race. They will ship Astra-level open-weight models. China is already good at this. Open weights become the default. That's a nightmare for US labs.

Bhavy☄️

26,050 次观看 • 19 天前

Did Theo just pull off a move that requires his brain? Who is really running this house? Is it a duo? Does this surprise me? Damn, Gabe, I thought you learned your lesson by underestimating Nia. Say it ain't so! Is the 'stake your claim thing' a good idea? Or should we be asking it differently...as in...is it a good idea to try, for the right person...at the right time?! All in all, this was a weird episode. But a good one. What did y'all think?

Did Theo just pull off a move that requires his brain? Who is really running this house? Is it a duo? Does this surprise me? Damn, Gabe, I thought you learned your lesson by underestimating Nia. Say it ain't so! Is the 'stake your claim thing' a good idea? Or should we be asking it differently...as in...is it a good idea to try, for the right person...at the right time?! All in all, this was a weird episode. But a good one. What did y'all think?

Wes Bergmann 🌋

25,543 次观看 • 11 个月前

I am splitting time between Godot & Unity but took a detour to see what the threejs hype was about. Was able to put together a 2.5D system pretty quickly with some of our 2D pipeline for a demo. Godot and threejs side-by-side same assets. Cave + Bridge scene is threejs. Idea on this one was a group of knights having to take back a world from evil with only their shields. All the swords flew up into the sky, suspended by something. Not in my usual style but really dig the lighting in threejs, scenes came together really quickly. The shield throw ended up going into the Moki demo as a leaf throw. Which one looks better?

I am splitting time between Godot & Unity but took a detour to see what the threejs hype was about. Was able to put together a 2.5D system pretty quickly with some of our 2D pipeline for a demo. Godot and threejs side-by-side same assets. Cave + Bridge scene is threejs. Idea on this one was a group of knights having to take back a world from evil with only their shields. All the swords flew up into the sky, suspended by something. Not in my usual style but really dig the lighting in threejs, scenes came together really quickly. The shield throw ended up going into the Moki demo as a leaf throw. Which one looks better?

Bruce✨

13,611 次观看 • 1 个月前

Perfume brands are paying agencies $50k for ads you can now make for under $2 😳 i tested it myself same prompt, two ai models. MiniMax H3 vs Seedance 2.5 seedance charged me 4x more. H3 still looked better what H3 got right: > the glass and lighting look real, not ai > brand name on the bottle came out clean, no melted text > 2k quality, 15 second videos > starts at $0.081 per second if you make ads, product videos or brand content, this is the cheapest way to look expensive both videos below, which one won? 👇

Isra

40,898 次观看 • 1 个月前

I tested Nutrient on a CMS-1500 health claim, which is exactly the kind of document that can trip up extraction systems. The red grid alone can get mangled by scanners, and the form is packed with dense fields where one wrong value can matter. What stood out to me was the grounding. Hover over any extracted field and Nutrient shows the exact place on the original claim where that value came from. So instead of asking “did the AI probably get this right?”, you can check the evidence yourself. For healthcare documents, that feels like a much better standard: not just extraction, but extraction you can actually verify.

I tested Nutrient on a CMS-1500 health claim, which is exactly the kind of document that can trip up extraction systems. The red grid alone can get mangled by scanners, and the form is packed with dense fields where one wrong value can matter. What stood out to me was the grounding. Hover over any extracted field and Nutrient shows the exact place on the original claim where that value came from. So instead of asking “did the AI probably get this right?”, you can check the evidence yourself. For healthcare documents, that feels like a much better standard: not just extraction, but extraction you can actually verify.

Rishabh

41,285 次观看 • 1 个月前

To me so far the big news with Grok 4.7 is it's speed. The new "fast" mode is actually extremely fast (idk what the TPS is but it's really good) It's very obviously not Astra/Fable class, SpaceXAi (or anyone else really) has yet to make one so far. Grok 4.7 is pretty clearly competing with Opus/Sol, and I would definitely take Grok 4.7 over either of them Price is very good, my gut says it's more efficient but I don't have hard numbers on that yet, and it's got all the main capabilities u need to actually be useful these days: subagents, long running tasks, looping, etc. Supposedly they trained it on the Grok Bot harness so it's really good at conversation + knowledge work. Really curious how that plays out, good product is SpaceXAi's biggest advantage right now so getting the models tailor made for those could work. Especially if that lets them keep the model on the cheaper side so they can serve more of it

To me so far the big news with Grok 4.7 is it's speed. The new "fast" mode is actually extremely fast (idk what the TPS is but it's really good) It's very obviously not Astra/Fable class, SpaceXAi (or anyone else really) has yet to make one so far. Grok 4.7 is pretty clearly competing with Opus/Sol, and I would definitely take Grok 4.7 over either of them Price is very good, my gut says it's more efficient but I don't have hard numbers on that yet, and it's got all the main capabilities u need to actually be useful these days: subagents, long running tasks, looping, etc. Supposedly they trained it on the Grok Bot harness so it's really good at conversation + knowledge work. Really curious how that plays out, good product is SpaceXAi's biggest advantage right now so getting the models tailor made for those could work. Especially if that lets them keep the model on the cheaper side so they can serve more of it

Ben Davis

54,012 次观看 • 11 天前

As someone who is pretty up to date in AI I’d say there are 2 times a year where I am genuinely blown away by something. A breakthrough that keeps me up at night. This is one of them. I mean the video of will smith genuinely freaks me out by how good it is. 2.5 years of AI progress and you think it’s hitting a wall. There’s no point in trying to convince people where this is going. The writing is on the wall. I can bring a horse to water but I can’t force him to drink.

As someone who is pretty up to date in AI I’d say there are 2 times a year where I am genuinely blown away by something. A breakthrough that keeps me up at night. This is one of them. I mean the video of will smith genuinely freaks me out by how good it is. 2.5 years of AI progress and you think it’s hitting a wall. There’s no point in trying to convince people where this is going. The writing is on the wall. I can bring a horse to water but I can’t force him to drink.

Chris

53,773 次观看 • 7 个月前

before pumpfun livestream feature is updated, before my account goes big and before I got some money, before everything, there was this token the beginning was so small at that time(over 2 years ago) rug was still rampant and I thought like 'why wouldn't they grow their project bigger rather than rugging at 10k?' sadly I became one of them now but regareless of it, I launched it just for 100% fun with buying 1 sol and turned on camera at TG group just for fun too someone said, "yo bro you should keep doing this this gonna be hella huge" so I did it I still remember the 2 guys who carried the whole project with max shilling and leading community members: Noble(this guy was pretty mean to me lol but still he was a goat) and Cassius(actual goat) it was a pure joy I did a stream 24/7 even while I was sleeping at TG and people were having fun in here(one girl took off her shirt when I was sleeping and I fucking missed it 💀) and my big bro Tyrelle Anderson-Brown came into my coin and helped me with 2 sol. I still remember this thankful money. with this I ate a nice dinner with my gf I still don't know the reason(maybe money laundaring?) but it went 5M and at this day when I woke up and checked my trojan, THIS WAS THE FIRST TIME I HIT 6 FIGS IN MY LIFE and I didn't sell a penny because community was more important than money at that time here's a list what community did for me - bought me a new iPhone - bought me a new MacBook(these two were for a better stream - my PC and phone was trash) - funded me almost every equipment for stream - formed a team with 10-11 members and kept supporting me they even put my sleeping video at Timesquare, NY here's a video so how can I sell this lmao but sadly the coin goes up, the coin goes down too and this happened to my coin too it was sooo tough days but I kept doing my best and it ended from up 130k to making 3k only this project was like if someone asks me "what did you do in this year? can you answer to this question with confidence?", I will answer this coin with 100% sure and I just turned on livestream with this coin just for fun too was very good days

before pumpfun livestream feature is updated, before my account goes big and before I got some money, before everything, there was this token the beginning was so small at that time(over 2 years ago) rug was still rampant and I thought like 'why wouldn't they grow their project bigger rather than rugging at 10k?' sadly I became one of them now but regareless of it, I launched it just for 100% fun with buying 1 sol and turned on camera at TG group just for fun too someone said, "yo bro you should keep doing this this gonna be hella huge" so I did it I still remember the 2 guys who carried the whole project with max shilling and leading community members: Noble(this guy was pretty mean to me lol but still he was a goat) and Cassius(actual goat) it was a pure joy I did a stream 24/7 even while I was sleeping at TG and people were having fun in here(one girl took off her shirt when I was sleeping and I fucking missed it 💀) and my big bro Tyrelle Anderson-Brown came into my coin and helped me with 2 sol. I still remember this thankful money. with this I ate a nice dinner with my gf I still don't know the reason(maybe money laundaring?) but it went 5M and at this day when I woke up and checked my trojan, THIS WAS THE FIRST TIME I HIT 6 FIGS IN MY LIFE and I didn't sell a penny because community was more important than money at that time here's a list what community did for me - bought me a new iPhone - bought me a new MacBook(these two were for a better stream - my PC and phone was trash) - funded me almost every equipment for stream - formed a team with 10-11 members and kept supporting me they even put my sleeping video at Timesquare, NY here's a video so how can I sell this lmao but sadly the coin goes up, the coin goes down too and this happened to my coin too it was sooo tough days but I kept doing my best and it ended from up 130k to making 3k only this project was like if someone asks me "what did you do in this year? can you answer to this question with confidence?", I will answer this coin with 100% sure and I just turned on livestream with this coin just for fun too was very good days

letterbomb 🟪🔶🟦⟠

14,212 次观看 • 5 个月前

I wanted to wait until the con is pretty much over. I reported this to Twitch Staff mid block party immediately, but there were souvenir switchblades available for purchase (or literally grab and walk out with) at one of the shops in the Block Party Gaslamp area. It was a whole collection with different names engraved so anybody could reach and grab it at any height. It was towards the back corner of the shop so you had to walk in and really browse to notice it. The shop itself was also at the very back corner so there was barely anybody browsing it. However when my group all saw this at the same time - we all had the same concerned look without even saying a word. We noticed these the 1st hour of the Party so I immediately wanted to leave.

I wanted to wait until the con is pretty much over. I reported this to Twitch Staff mid block party immediately, but there were souvenir switchblades available for purchase (or literally grab and walk out with) at one of the shops in the Block Party Gaslamp area. It was a whole collection with different names engraved so anybody could reach and grab it at any height. It was towards the back corner of the shop so you had to walk in and really browse to notice it. The shop itself was also at the very back corner so there was barely anybody browsing it. However when my group all saw this at the same time - we all had the same concerned look without even saying a word. We noticed these the 1st hour of the Party so I immediately wanted to leave.

⋆🌙 ⋆ CTRL+F:『🐇』🎧🔜 OFFKAI ANIMERAVE.EXE

2,025,404 次观看 • 11 个月前

THE CLOUD BILL WAS $14,000 A MONTH. EVERY MONTH. JUST FOR GPU ACCESS. No ownership. No hardware. Just renting someone else's chips and watching the invoice grow. So this startup did the math and bought 1,000 Mac Mini M4s instead. $599 each. One-time. Total: ~$599,000 upfront. Sounds insane. Until you do the math the other way. $14K/mo in cloud GPUs is $168K a year. In 3.5 years you've burned through $599K and own nothing. The meter just keeps running. These guys own the hardware. Same output. Less power draw. And after year 4, every month of compute is basically free. The M4 was never built for data centers. But its performance per watt is so good that someone looked at a rack of a thousand of them and thought "why not." That's when it clicks. The AI infrastructure race isn't about who has the biggest GPU anymore. It's about who figured out they were overpaying for one. Would you make this switch?

THE CLOUD BILL WAS $14,000 A MONTH. EVERY MONTH. JUST FOR GPU ACCESS. No ownership. No hardware. Just renting someone else's chips and watching the invoice grow. So this startup did the math and bought 1,000 Mac Mini M4s instead. $599 each. One-time. Total: ~$599,000 upfront. Sounds insane. Until you do the math the other way. $14K/mo in cloud GPUs is $168K a year. In 3.5 years you've burned through $599K and own nothing. The meter just keeps running. These guys own the hardware. Same output. Less power draw. And after year 4, every month of compute is basically free. The M4 was never built for data centers. But its performance per watt is so good that someone looked at a rack of a thousand of them and thought "why not." That's when it clicks. The AI infrastructure race isn't about who has the biggest GPU anymore. It's about who figured out they were overpaying for one. Would you make this switch?

Framez

18,004 次观看 • 1 个月前

AI has had exactly two scaling axes that worked so far, and the second one is starting to look finite too the first one was pretraining: with scaling parameters and data, we got world knowledge (i.e. ChatGPT had read enough to know things), but it started saturating a while ago the second one was RL, and people had been doing RL the whole time before that: RLHF is RL but it never scaled far because it was trying to control the exact output, which tokens come out, how the text reads, but you can only push that so far before you’re just polishing RLVR dropped that constraint: giving the model a task, then checking whether the final answer is right, and ignoring everything in between -- so the model does whatever it wants in the middle and only the endpoint gets graded, and that’s much closer to actual RL and it’s what bought us planning and reasoning (arguably, tool use sits around 2.5 on this list -- while useful, it's not a different kind of thing) so one axis gave knowledge, the other gave reasoning, and both of them are one model working alone the next axis is how many models you can get working on the same problem, which is a different kind of axis than the previous two we know that multi-agent RL has always been the harder problem: I spent years in that literature and the gap between single-agent and multi-agent is definitely not incremental -- it’s a whole different class of difficulty! which is also why the derivatives are steep at the start, nobody has picked the easy wins yet... and the thing that gates this multi-agent coordination is communication: models can only coordinate as well as they can exchange information, and right now they do that by writing sentences to each other imagine what could we possibly achieve if we properly open that third axis development by letting models to exchange information in their native "language" without loosing any computational data that they produce during inference

AI has had exactly two scaling axes that worked so far, and the second one is starting to look finite too the first one was pretraining: with scaling parameters and data, we got world knowledge (i.e. ChatGPT had read enough to know things), but it started saturating a while ago the second one was RL, and people had been doing RL the whole time before that: RLHF is RL but it never scaled far because it was trying to control the exact output, which tokens come out, how the text reads, but you can only push that so far before you’re just polishing RLVR dropped that constraint: giving the model a task, then checking whether the final answer is right, and ignoring everything in between -- so the model does whatever it wants in the middle and only the endpoint gets graded, and that’s much closer to actual RL and it’s what bought us planning and reasoning (arguably, tool use sits around 2.5 on this list -- while useful, it's not a different kind of thing) so one axis gave knowledge, the other gave reasoning, and both of them are one model working alone the next axis is how many models you can get working on the same problem, which is a different kind of axis than the previous two we know that multi-agent RL has always been the harder problem: I spent years in that literature and the gap between single-agent and multi-agent is definitely not incremental -- it’s a whole different class of difficulty! which is also why the derivatives are steep at the start, nobody has picked the easy wins yet... and the thing that gates this multi-agent coordination is communication: models can only coordinate as well as they can exchange information, and right now they do that by writing sentences to each other imagine what could we possibly achieve if we properly open that third axis development by letting models to exchange information in their native "language" without loosing any computational data that they produce during inference

Sasha Malysheva

15,053 次观看 • 1 个月前

Astra (GPT-6) is here!!! I've had early access and tested it like crazy with things like games, code, writing, browser control, presentations and general knowledge work. This is the best model I've ever used. Period. (Incredible demos below in this thread ⬇️) Here's my take on Astra: > It's insanely capable. This feels like a massive improvement, not just an incremental change. This is especially true with zero-shot prompts. > It's all about knowledge work. Slide creation, analysis, writing, and browser control. And oh my...it's so good at browser control. GPT-5.6 was already fantastic at doing things in the browser, Astra is another level and significantly faster. > We're closer than ever (arrived?) at prompt-to-playable game. And I don't just mean only playable, these are actually fun games. I bet if someone with a great eye for games used Astra, they could create a viral game within 1-2 weeks. > Astra is better at writing but not perfect. It removes much of the "AI Smell" we're all familiar with but some stink still survived. > It has a tendency to use the same design colors and look/feel as GPT-5.6 (forrest green anyone?) but it is more steerable in design than previous models. > It's highly steerable in general. A little nudge goes a long way. When I first started using Astra, almost every task I gave it would go for ~30 minutes. I wanted it to keep working. Adding more specifics to a prompt helped greatly with it's ability to work for a long time. > Astra's 3D understanding is unmatched. 3D asset creation was consistent and easy and its spacial awareness while building complex 3D worlds blew me away. I'm still getting familiar with Astra but this will now be my go-to model for any difficult work I have. Check out the demos below: 👇

Astra (GPT-6) is here!!! I've had early access and tested it like crazy with things like games, code, writing, browser control, presentations and general knowledge work. This is the best model I've ever used. Period. (Incredible demos below in this thread ⬇️) Here's my take on Astra: > It's insanely capable. This feels like a massive improvement, not just an incremental change. This is especially true with zero-shot prompts. > It's all about knowledge work. Slide creation, analysis, writing, and browser control. And oh my...it's so good at browser control. GPT-5.6 was already fantastic at doing things in the browser, Astra is another level and significantly faster. > We're closer than ever (arrived?) at prompt-to-playable game. And I don't just mean only playable, these are actually fun games. I bet if someone with a great eye for games used Astra, they could create a viral game within 1-2 weeks. > Astra is better at writing but not perfect. It removes much of the "AI Smell" we're all familiar with but some stink still survived. > It has a tendency to use the same design colors and look/feel as GPT-5.6 (forrest green anyone?) but it is more steerable in design than previous models. > It's highly steerable in general. A little nudge goes a long way. When I first started using Astra, almost every task I gave it would go for ~30 minutes. I wanted it to keep working. Adding more specifics to a prompt helped greatly with it's ability to work for a long time. > Astra's 3D understanding is unmatched. 3D asset creation was consistent and easy and its spacial awareness while building complex 3D worlds blew me away. I'm still getting familiar with Astra but this will now be my go-to model for any difficult work I have. Check out the demos below: 👇

Matthew Berman

1,907,120 次观看 • 29 天前

zb1 is trending on theqoo HOT category, where over 200+ comments are praising zb1’s vocals, dance, everything, especially their cover of “sherlock (clue + note)” by shinee “damn the determination is insane, even with just five members the stage still feels full” “honestly, this was the most memorable stage for me at kcon today lol. the nerve to cover “sherlock” live is insane, they did so well. i didn’t realize zb1 were this talented” “sherlock is such a difficult song, so honestly i’m shocked a 5th gen group can pull this off. they’re seriously good” “the determination was crazy. this song is really hard and keeps repeating high notes… and the dancing was perfectly in sync too” “just deciding to cover sherlock already shows legendary determination lol but they seriously did so well” “since this is their first stage after reorganizing into five members, i love that you can feel the determination right from the song choice. they did well” “i guess because the current members are all main vocal or strong vocal members, they’re insanely good” “sherlock is insanely high but hearing them do it live so powerfully feels so satisfying. zb1 are insanely good” “it’s been a while since i’ve seen a stage this full of determination, my dopamine is going crazy” “i usually don’t watch other idols’ stages but the determination was insane so i got totally focused. they’re seriously good” “wait this sounds fully live though? sung hanbin is surprising, i thought he was only good at dancing but he sings well too??”

zb1 is trending on theqoo HOT category, where over 200+ comments are praising zb1’s vocals, dance, everything, especially their cover of “sherlock (clue + note)” by shinee “damn the determination is insane, even with just five members the stage still feels full” “honestly, this was the most memorable stage for me at kcon today lol. the nerve to cover “sherlock” live is insane, they did so well. i didn’t realize zb1 were this talented” “sherlock is such a difficult song, so honestly i’m shocked a 5th gen group can pull this off. they’re seriously good” “the determination was crazy. this song is really hard and keeps repeating high notes… and the dancing was perfectly in sync too” “just deciding to cover sherlock already shows legendary determination lol but they seriously did so well” “since this is their first stage after reorganizing into five members, i love that you can feel the determination right from the song choice. they did well” “i guess because the current members are all main vocal or strong vocal members, they’re insanely good” “sherlock is insanely high but hearing them do it live so powerfully feels so satisfying. zb1 are insanely good” “it’s been a while since i’ve seen a stage this full of determination, my dopamine is going crazy” “i usually don’t watch other idols’ stages but the determination was insane so i got totally focused. they’re seriously good” “wait this sounds fully live though? sung hanbin is surprising, i thought he was only good at dancing but he sings well too??”

☁️ (IA)

15,880 次观看 • 4 个月前

Right now, you may not have access to models like GPT‑5.6 Sol, GPT‑4.6 Terra, GPT‑5.6 Luna, Claude Mythos 5, or Claude Fable 5. But you can run something surprisingly powerful today, locally, and completely free. in the next 10 mins on your 8 GB VRAM gaming laptop. Gemma 4 26B A4B QAT (MoE) delivers strong performance on a standard 8 GB VRAM GPU using Ollama, with no API, no usage limits, and no external dependencies. Out of the box, it reaches around 20 tokens per second without any optimizations. Only one command in your terminal: Ollama run gemma4:26b This means: Full offline capability (privacy by default) Zero recurring cost Competitive performance for many real world tasks Fast enough for interactive use on cheap consumer hardware If you're waiting for cutting edge cloud models, you're missing what is already practical today: a capable, local LLM that runs entirely on your own machine.

Right now, you may not have access to models like GPT‑5.6 Sol, GPT‑4.6 Terra, GPT‑5.6 Luna, Claude Mythos 5, or Claude Fable 5. But you can run something surprisingly powerful today, locally, and completely free. in the next 10 mins on your 8 GB VRAM gaming laptop. Gemma 4 26B A4B QAT (MoE) delivers strong performance on a standard 8 GB VRAM GPU using Ollama, with no API, no usage limits, and no external dependencies. Out of the box, it reaches around 20 tokens per second without any optimizations. Only one command in your terminal: Ollama run gemma4:26b This means: Full offline capability (privacy by default) Zero recurring cost Competitive performance for many real world tasks Fast enough for interactive use on cheap consumer hardware If you're waiting for cutting edge cloud models, you're missing what is already practical today: a capable, local LLM that runs entirely on your own machine.

Alok

65,387 次观看 • 3 个月前

A lesson for every Polymarket bot developer: I built a strategy that looked perfect on paper. Backtested it. Looked like a winner. Almost went live. Then i actually measured real costs. Strategy was dead before the first trade. And this is what bot building on Polymarket actually looks like. Here is what happened (and what you MUST know): Backtested mean reversion on crypto dips. SOL came back at +44% return and 70.5% win rate. Beautiful clean curve. Looked ready to ship. Then i measured real round-trip costs on SOL flash dips. Backtest assumed 0.45% in fees and slippage. Reality was 1.44%. Strategy stops working at 0.70%. Starting again. But the lesson was worth more than any profit the strategy could have made. Here is what i actually learned: Taking a dip with a market order means you eat the spread the dip just created. The volatility making your signal is the same volatility destroying your fill. You see the opportunity. You enter. You already lost. But resting a limit order below market and letting the dip come to you? You collect the maker rebate instead. Same thesis. Completely opposite execution. One bleeds money, one prints it. That one realization changed how i think about bot strategy entirely. 180 strategies tested to get there. 179 dead. That is not failure. That is how you find the 5 that actually work. Building a bot on Polymarket is not about finding a magic strategy. It is about eliminating every wrong answer until only the right one is left.

A lesson for every Polymarket bot developer: I built a strategy that looked perfect on paper. Backtested it. Looked like a winner. Almost went live. Then i actually measured real costs. Strategy was dead before the first trade. And this is what bot building on Polymarket actually looks like. Here is what happened (and what you MUST know): Backtested mean reversion on crypto dips. SOL came back at +44% return and 70.5% win rate. Beautiful clean curve. Looked ready to ship. Then i measured real round-trip costs on SOL flash dips. Backtest assumed 0.45% in fees and slippage. Reality was 1.44%. Strategy stops working at 0.70%. Starting again. But the lesson was worth more than any profit the strategy could have made. Here is what i actually learned: Taking a dip with a market order means you eat the spread the dip just created. The volatility making your signal is the same volatility destroying your fill. You see the opportunity. You enter. You already lost. But resting a limit order below market and letting the dip come to you? You collect the maker rebate instead. Same thesis. Completely opposite execution. One bleeds money, one prints it. That one realization changed how i think about bot strategy entirely. 180 strategies tested to get there. 179 dead. That is not failure. That is how you find the 5 that actually work. Building a bot on Polymarket is not about finding a magic strategy. It is about eliminating every wrong answer until only the right one is left.

Oracle Boar

14,379 次观看 • 5 个月前

I’m late to this, but I see both sides of the coin here. On one, of course Kyle Fletcher’s gonna be bummed by getting screwed out of his big stadium moment in front of his family while his anxiety levels were at an all-time high. On the other, I wish he included more context as to why that happened and what happened to Adam Cole, as well as the fact that he may never wrestle again and how that was announced on the day of All In, because it makes complete sense as to why that happened when you take everything into account. Also, I can’t believe I’m saying this again but please walk with me on this one: ADAM. COLE. DID. NOT. GET. HURT. ON. THAT. FUCKING. POWERBOMB. YOU. IMBECILES. Watch his arms hit the fucking apron as his head hits nothing. You can say it was a dumb spot all you want, that’s not why Adam Cole is sitting at home right now.

I’m late to this, but I see both sides of the coin here. On one, of course Kyle Fletcher’s gonna be bummed by getting screwed out of his big stadium moment in front of his family while his anxiety levels were at an all-time high. On the other, I wish he included more context as to why that happened and what happened to Adam Cole, as well as the fact that he may never wrestle again and how that was announced on the day of All In, because it makes complete sense as to why that happened when you take everything into account. Also, I can’t believe I’m saying this again but please walk with me on this one: ADAM. COLE. DID. NOT. GET. HURT. ON. THAT. FUCKING. POWERBOMB. YOU. IMBECILES. Watch his arms hit the fucking apron as his head hits nothing. You can say it was a dumb spot all you want, that’s not why Adam Cole is sitting at home right now.

KLA Clips/KnicksLoseAgain

320,855 次观看 • 6 个月前