Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

THIS AI DEVELOPER RUNS 120B PARAMETER MODELS LOCALLY AND HIS MACBOOK PRO CAN’T EVEN OPEN THEM he bought a DGX Spark, gave it the model, went back to his desk by end of day the inference was done, tested, committed his team is still waiting for the cloud GPU... show more

leopardracer

9,040 subscribers

83,303 views • 14 days ago •via X (Twitter)

Education News & Politics Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

JENSEN HUANG UNVEILED A $3,000 SUPERCOMPUTER THAT REPLACES $1,900/MONTH IN CLOUD GPUS WITH $2 IN ELECTRICITY. IT FITS ON YOUR DESK a developer was paying $1,900 a month renting a100 and h100 cloud gpus. he bought the dgx spark, set it up in an afternoon, and his monthly bill dropped from $1,900 to $2 in electricity the box runs 128gb of unified memory and 1 petaflop of ai compute. a 4090 gives you 24gb of vram and taps out at 30b models. the spark loads 70b at full precision and stretches to 200b without breaking a sweat break-even hit in 6 weeks. after that the $1,890 he used to wire to a rental company every month stayed in his business. first year that's $22,000 back install ollama, change one line in your code, point it at localhost instead of the cloud. nothing else changes. nothing leaves your machine and nothing costs money per request cloud gpu costs aren't getting cheaper and rate limits keep getting tighter. the people who pulled their ai workloads onto a box on their desk in 2026 are going to look very far ahead in 2028 bookmark this and read the article below

JENSEN HUANG UNVEILED A $3,000 SUPERCOMPUTER THAT REPLACES $1,900/MONTH IN CLOUD GPUS WITH $2 IN ELECTRICITY. IT FITS ON YOUR DESK a developer was paying $1,900 a month renting a100 and h100 cloud gpus. he bought the dgx spark, set it up in an afternoon, and his monthly bill dropped from $1,900 to $2 in electricity the box runs 128gb of unified memory and 1 petaflop of ai compute. a 4090 gives you 24gb of vram and taps out at 30b models. the spark loads 70b at full precision and stretches to 200b without breaking a sweat break-even hit in 6 weeks. after that the $1,890 he used to wire to a rental company every month stayed in his business. first year that's $22,000 back install ollama, change one line in your code, point it at localhost instead of the cloud. nothing else changes. nothing leaves your machine and nothing costs money per request cloud gpu costs aren't getting cheaper and rate limits keep getting tighter. the people who pulled their ai workloads onto a box on their desk in 2026 are going to look very far ahead in 2028 bookmark this and read the article below

starmex

140,989 views • 15 days ago

NVIDIA will refund your cloud bill if you let them put a $250,000 supercomputer on your desk for $2,999 The $2,999 DGX Spark puts a 128GB AI supercomputer on your desk the same 70B models you rented in the cloud, but nothing goes over the network and no ToS governs the machine you own $22,000 in annual savings is the big number everyone posts The quiet number is even bigger it’s the contracts you’ve stopped losing due to data residency Owning your own compute is no longer just budget optimization it’s a way to sell to customers who can’t send their data to the cloud

NVIDIA will refund your cloud bill if you let them put a $250,000 supercomputer on your desk for $2,999 The $2,999 DGX Spark puts a 128GB AI supercomputer on your desk the same 70B models you rented in the cloud, but nothing goes over the network and no ToS governs the machine you own $22,000 in annual savings is the big number everyone posts The quiet number is even bigger it’s the contracts you’ve stopped losing due to data residency Owning your own compute is no longer just budget optimization it’s a way to sell to customers who can’t send their data to the cloud

Asteri

143,167 views • 12 days ago

people burn $1,900 a month on cloud, this $2,999 NVIDIA box runs 5 AI models at once and costs $10 to power 128GB of unified memory, stack two together and you double it, enough to run models a $2,000 graphics card cannot even open in the demo it ran Qwen3 Vision, Nemotron, a voice model, text to speech and speech to text all at once, and only used half the box here is the math that makes people switch: > cloud GPUs were costing $1,900 every single month, billed forever > the box is $2,999 one time, then about $10 of electricity a month > it paid for itself in two months, $22,000 stayed in the business in year one > zero data ever leaves your desk, no rental, no rate limits most people are still renting what this owns outright the full breakdown of the box and the exact math is in the article below

people burn $1,900 a month on cloud, this $2,999 NVIDIA box runs 5 AI models at once and costs $10 to power 128GB of unified memory, stack two together and you double it, enough to run models a $2,000 graphics card cannot even open in the demo it ran Qwen3 Vision, Nemotron, a voice model, text to speech and speech to text all at once, and only used half the box here is the math that makes people switch: > cloud GPUs were costing $1,900 every single month, billed forever > the box is $2,999 one time, then about $10 of electricity a month > it paid for itself in two months, $22,000 stayed in the business in year one > zero data ever leaves your desk, no rental, no rate limits most people are still renting what this owns outright the full breakdown of the box and the exact math is in the article below

John Doe

30,995 views • 14 days ago

NVIDIA will refund your cloud GPU bill if you let them bolt a $250,000 AI supercomputer to your desk for $2,999. NVIDIA's $2,999 DGX Spark puts a 128GB AI supercomputer on your desk runs the same 70B models you've been renting in the cloud, but nothing crosses your network and no ToS governs a machine you own. the $22K/year savings everyone's posting is the loud number. the quiet one is bigger: the contracts you stopped losing on data residency. own-compute isn't a budget optimization anymore. it's a sales motion. ↓ builder benchmark from the week the box shipped. real numbers, not marketing.

NVIDIA will refund your cloud GPU bill if you let them bolt a $250,000 AI supercomputer to your desk for $2,999. NVIDIA's $2,999 DGX Spark puts a 128GB AI supercomputer on your desk runs the same 70B models you've been renting in the cloud, but nothing crosses your network and no ToS governs a machine you own. the $22K/year savings everyone's posting is the loud number. the quiet one is bigger: the contracts you stopped losing on data residency. own-compute isn't a budget optimization anymore. it's a sales motion. ↓ builder benchmark from the week the box shipped. real numbers, not marketing.

dunik

817,961 views • 16 days ago

I tested Gemma 4 12b on my Macbook Pro M5 Max with 128GB of unified memory. It runs VERY slow for a 12b parameter model. 44 tokens/second. The lava lamp that it produced was also terrible. I would stick to Qwen 3.6 27b or Qwen 3.6 35b. Google is still way behind, even with their local models.

I tested Gemma 4 12b on my Macbook Pro M5 Max with 128GB of unified memory. It runs VERY slow for a 12b parameter model. 44 tokens/second. The lava lamp that it produced was also terrible. I would stick to Qwen 3.6 27b or Qwen 3.6 35b. Google is still way behind, even with their local models.

BridgeMind

69,621 views • 9 days ago

✨ Personal AI is the next computing platform. AI is shifting from something you access to something you build with, locally, at the edge, and across systems. We’re unlocking new possibilities for developers: • AMD Ryzen AI Halo, a local-first developer system, preorder starting in June, develop AI without limits on your desk • Gorgon Halo with up to 192GB unified memory, supporting 300B+ parameter models locally, run massive models locally We’re excited to partner with clem 🤗🤗, Co-founder and CEO of Hugging Face, to advance open-source AI for Ryzen AI. Our focus is seamless AI, from model to deployment. Cloud, edge, device. One continuum. Multi-agent systems, local inference at scale, open models as infrastructure. This is the next computing era 🚀

✨ Personal AI is the next computing platform. AI is shifting from something you access to something you build with, locally, at the edge, and across systems. We’re unlocking new possibilities for developers: • AMD Ryzen AI Halo, a local-first developer system, preorder starting in June, develop AI without limits on your desk • Gorgon Halo with up to 192GB unified memory, supporting 300B+ parameter models locally, run massive models locally We’re excited to partner with clem 🤗🤗, Co-founder and CEO of Hugging Face, to advance open-source AI for Ryzen AI. Our focus is seamless AI, from model to deployment. Cloud, edge, device. One continuum. Multi-agent systems, local inference at scale, open models as infrastructure. This is the next computing era 🚀

Jack Huynh

47,062 views • 25 days ago

THEY RAN A LOCAL AI MODEL ON A RASPBERRY PI THAT COSTS LESS THAN $300 no cloud no subscription no data center device smaller than your wallet output on a 27 year old iMac I tested local ai this week on a macbook and a phone the hardware keeps getting cheaper and the models keep getting smarter bookmark this if you’re still paying $20 a month for something your $300 raspberry pi can do

THEY RAN A LOCAL AI MODEL ON A RASPBERRY PI THAT COSTS LESS THAN $300 no cloud no subscription no data center device smaller than your wallet output on a 27 year old iMac I tested local ai this week on a macbook and a phone the hardware keeps getting cheaper and the models keep getting smarter bookmark this if you’re still paying $20 a month for something your $300 raspberry pi can do

leopardracer

73,534 views • 24 days ago

This Chinese developer linked two $2,999 NVIDIA DGX Sparks into one box and runs the full Qwen3-235B at home, after dropping his $1,999-a-month cloud bill to zero. He wired 2 small boxes into a single computer, split a giant 235-billion-parameter model in half between them, and serves it across his own network at about 10 tokens a second, with no internet, no cloud, right there on the desk. No data center, no thousand-dollar graphics cards, no monthly cloud bill. Just him, 2 gold boxes the size of a sandwich, one cable between them, and 1 power strip. And here is the whole payoff. He used to pay the cloud $1,999 a month for the same model, and the meter ticked on every request. Now he paid $5,998 once for 2 boxes, they covered their cost in 3 months, and after that he sends as many requests as he wants for free, only electricity. The two Sparks talk over one fast cable, each holds 128GB of memory, and together they carry the whole model, about 73GB loaded per box, with the chip inside pinned near the limit at 96%. Both boxes work as one and keep trading data over the cable, with no cloud in the loop and no single word leaking out. The ready model sits on one local address, and any app on his network calls it as easily as ChatGPT. And here is how he described, in plain words, what this pair of boxes does: "this is a pair of boxes that holds the huge Qwen3-235B model and serves it to one network. the model is split in half, and each box owns its half. parts: // Box 1 (holds the first half of the model and starts the answer fast, the first word appears in under a second) // Box 2 (holds the second half and writes out the rest, about 10 tokens a second) // Cable (connects the 2 boxes and moves data between them on every step, with no lag) // Address (one local address where any app sends its request, like to a cloud model) // Test (a script that runs big prompts through and measures speed and delays) // Monitor (checks temperature, power draw, and load on both boxes every 2 seconds). the model never goes to the cloud. he only steps in when a box runs hotter than 80 degrees or the cable between them starts dropping data." So the system knows exactly what it is, what it is for, and where its limits are. It knows it has to hold the whole huge model across 2 boxes on its own. It knows it has to answer every request locally, with no meter, no limits, and no internet. It knows the human is only needed when a box overheats or the link between them stalls. → The setup runs around the clock on 2 boxes, each pulling under 60 watts → However many requests he sends, the monthly bill is $0, only electricity → The first box starts the answer in under a second → The second writes text at about 10 tokens a second → One request at a time: 838 tokens in 85 seconds, first word in 0.8s → Two requests at once: 697 tokens in 108 seconds, first word in 0.7s → Both boxes sit at 96% load and warm up to 76-78 degrees And only when a chip in a box runs hotter than 80 degrees or the cable between the 2 Sparks drops data does the system call the owner. And when he himself is out on a run or in a coffee shop, he still reaches his own model at home from his phone: sends a big prompt to the local Qwen3-235B, gets the full answer back in under a minute and a half, with no token meter ticking and no limit to hit. Here is what the test shows on his screen during one of the night runs: "one request at a time: 838 tokens in 84.9 seconds, first word in 0.8s, then 0.1s per token." "two requests at once: 697 tokens in 107.6 seconds, first word in 0.7s, then 0.15s per token." "Box 1: chip at 96% load, 76 degrees, 56 watts, 73GB used in memory." "Box 2: chip at 96% load, 78 degrees, 56 watts, the Qwen3-235B model fully loaded." And while everyone around is paying for AI by the month and bumping into limits, his top-tier model just sits on the desk and works as much as he wants: his own little power plant instead of a forever meter. He has no server rack of his own and no cloud account behind it. Just 2 DGX Spark boxes on a desk, one model split in half between them, one local address, and a folder of prompts next to it. Out of everything I have seen this year, this is the cleanest way to stop paying for AI: $5,998 of hardware on the desk once, $0 a month to the cloud, unlimited forever, and between them 2 gold boxes, 1 cable, and the full Qwen3-235B answering at home with no internet.

This Chinese developer linked two $2,999 NVIDIA DGX Sparks into one box and runs the full Qwen3-235B at home, after dropping his $1,999-a-month cloud bill to zero. He wired 2 small boxes into a single computer, split a giant 235-billion-parameter model in half between them, and serves it across his own network at about 10 tokens a second, with no internet, no cloud, right there on the desk. No data center, no thousand-dollar graphics cards, no monthly cloud bill. Just him, 2 gold boxes the size of a sandwich, one cable between them, and 1 power strip. And here is the whole payoff. He used to pay the cloud $1,999 a month for the same model, and the meter ticked on every request. Now he paid $5,998 once for 2 boxes, they covered their cost in 3 months, and after that he sends as many requests as he wants for free, only electricity. The two Sparks talk over one fast cable, each holds 128GB of memory, and together they carry the whole model, about 73GB loaded per box, with the chip inside pinned near the limit at 96%. Both boxes work as one and keep trading data over the cable, with no cloud in the loop and no single word leaking out. The ready model sits on one local address, and any app on his network calls it as easily as ChatGPT. And here is how he described, in plain words, what this pair of boxes does: "this is a pair of boxes that holds the huge Qwen3-235B model and serves it to one network. the model is split in half, and each box owns its half. parts: // Box 1 (holds the first half of the model and starts the answer fast, the first word appears in under a second) // Box 2 (holds the second half and writes out the rest, about 10 tokens a second) // Cable (connects the 2 boxes and moves data between them on every step, with no lag) // Address (one local address where any app sends its request, like to a cloud model) // Test (a script that runs big prompts through and measures speed and delays) // Monitor (checks temperature, power draw, and load on both boxes every 2 seconds). the model never goes to the cloud. he only steps in when a box runs hotter than 80 degrees or the cable between them starts dropping data." So the system knows exactly what it is, what it is for, and where its limits are. It knows it has to hold the whole huge model across 2 boxes on its own. It knows it has to answer every request locally, with no meter, no limits, and no internet. It knows the human is only needed when a box overheats or the link between them stalls. → The setup runs around the clock on 2 boxes, each pulling under 60 watts → However many requests he sends, the monthly bill is $0, only electricity → The first box starts the answer in under a second → The second writes text at about 10 tokens a second → One request at a time: 838 tokens in 85 seconds, first word in 0.8s → Two requests at once: 697 tokens in 108 seconds, first word in 0.7s → Both boxes sit at 96% load and warm up to 76-78 degrees And only when a chip in a box runs hotter than 80 degrees or the cable between the 2 Sparks drops data does the system call the owner. And when he himself is out on a run or in a coffee shop, he still reaches his own model at home from his phone: sends a big prompt to the local Qwen3-235B, gets the full answer back in under a minute and a half, with no token meter ticking and no limit to hit. Here is what the test shows on his screen during one of the night runs: "one request at a time: 838 tokens in 84.9 seconds, first word in 0.8s, then 0.1s per token." "two requests at once: 697 tokens in 107.6 seconds, first word in 0.7s, then 0.15s per token." "Box 1: chip at 96% load, 76 degrees, 56 watts, 73GB used in memory." "Box 2: chip at 96% load, 78 degrees, 56 watts, the Qwen3-235B model fully loaded." And while everyone around is paying for AI by the month and bumping into limits, his top-tier model just sits on the desk and works as much as he wants: his own little power plant instead of a forever meter. He has no server rack of his own and no cloud account behind it. Just 2 DGX Spark boxes on a desk, one model split in half between them, one local address, and a folder of prompts next to it. Out of everything I have seen this year, this is the cleanest way to stop paying for AI: $5,998 of hardware on the desk once, $0 a month to the cloud, unlimited forever, and between them 2 gold boxes, 1 cable, and the full Qwen3-235B answering at home with no internet.

Blaze

93,219 views • 15 days ago

THIS DEVELOPER HASN’T PAID AN API BILL IN 3 MONTHS. HIS AGENTS RAN 10,000 TIMES FOR FREE he built a local AI lab under his desk two GPUs 32GB VRAM zero rate limits his agents loop 400 times if they want to his coworkers are still watching the usage dashboard the only thing separating them: llama.cpp + llama-swap every prompt stays on his machine every experiment costs $0 zero invoices bookmark & like this before your next API bill hits

THIS DEVELOPER HASN’T PAID AN API BILL IN 3 MONTHS. HIS AGENTS RAN 10,000 TIMES FOR FREE he built a local AI lab under his desk two GPUs 32GB VRAM zero rate limits his agents loop 400 times if they want to his coworkers are still watching the usage dashboard the only thing separating them: llama.cpp + llama-swap every prompt stays on his machine every experiment costs $0 zero invoices bookmark & like this before your next API bill hits

leopardracer

539,011 views • 1 month ago

Nvidia will now let regular operators replace $25,200/year of cloud GPUs with two boxes on a desk DGX Spark just shipped, and the math is brutal: → One box ($2,999) runs the same workload AWS charges $1,900/month for → Two boxes ($3,248 total) replace ChatGPT, Claude, Cursor, OpenAI API and cloud H100s → Paid back in 6 weeks → +$8,400 in pure margin by month 4 → $25,200/year off the subscription bill, forever Tier 1 - $249, size of a deck of cards: → summarization, voice, classification, RAG over private docs Tier 2 - $2,999, fits on a desk: → 70B at full BF16, 405B at Q3, fine-tuning, long-context inference Stack 4x Tier 2 → run a 1.6 trillion parameter model on your desk for under $12K Territory that used to require six-figure cloud commitments Both boxes speak OpenAI-compatible API → migration is literally 1 line of code Save this. Your cloud bill is about to look optional. Full breakdown 👇

Nvidia will now let regular operators replace $25,200/year of cloud GPUs with two boxes on a desk DGX Spark just shipped, and the math is brutal: → One box ($2,999) runs the same workload AWS charges $1,900/month for → Two boxes ($3,248 total) replace ChatGPT, Claude, Cursor, OpenAI API and cloud H100s → Paid back in 6 weeks → +$8,400 in pure margin by month 4 → $25,200/year off the subscription bill, forever Tier 1 - $249, size of a deck of cards: → summarization, voice, classification, RAG over private docs Tier 2 - $2,999, fits on a desk: → 70B at full BF16, 405B at Q3, fine-tuning, long-context inference Stack 4x Tier 2 → run a 1.6 trillion parameter model on your desk for under $12K Territory that used to require six-figure cloud commitments Both boxes speak OpenAI-compatible API → migration is literally 1 line of code Save this. Your cloud bill is about to look optional. Full breakdown 👇

ZEUS⚡️

14,199 views • 15 days ago

This guy plugged a DGX Spark (the $3K Nvidia box) and a Mac Mini M4 together to run AI and what happened next surprised everyone > the Nvidia box handles the hard part - processing your prompt in milliseconds > Mac Mini M4 handles the fast part - generating the response at memory bandwidth speeds nothing else can match > together they hit 84 tokens per second on Llama - 6x faster than the Spark alone > running compute agents locally on this setup means your data never leaves your hardware > two boxes. two different architectures. one AI system that Deepseek runs at data center scale he ran it on his desk save this. the way we build local AI is about to change

This guy plugged a DGX Spark (the $3K Nvidia box) and a Mac Mini M4 together to run AI and what happened next surprised everyone > the Nvidia box handles the hard part - processing your prompt in milliseconds > Mac Mini M4 handles the fast part - generating the response at memory bandwidth speeds nothing else can match > together they hit 84 tokens per second on Llama - 6x faster than the Spark alone > running compute agents locally on this setup means your data never leaves your hardware > two boxes. two different architectures. one AI system that Deepseek runs at data center scale he ran it on his desk save this. the way we build local AI is about to change

Mr. Buzzoni

152,897 views • 16 days ago

8. DGX Spark = personal AI supercomputer Imagine training your own LLM at your desk. DGX Spark is the dev-friendly AI rig built for workstation scale. → Shipping via ASUS, Dell, MSI → Trains large models locally → No server farm needed

8. DGX Spark = personal AI supercomputer Imagine training your own LLM at your desk. DGX Spark is the dev-friendly AI rig built for workstation scale. → Shipping via ASUS, Dell, MSI → Trains large models locally → No server farm needed

Shruti

35,907 views • 1 year ago

This guy runs an Nvidia Spark from his desk and pulls in $8,000 a month The device is the size of a small speaker and sits right on his desk. Inside is a full Nvidia GPU running AI workloads 24/7. He rents out the compute power to companies that need local AI processing. They pay him a flat monthly fee to use it remotely. $8,000 lands in his account every single month. He just plugged it in and that was it. The $4,000 box paid for itself in two weeks. Now it prints $96,000 a year from a device smaller than a coffee machine. Companies are desperate for local compute and regular people are quietly supplying it. Save this before everyone figures out what he already knows.

This guy runs an Nvidia Spark from his desk and pulls in $8,000 a month The device is the size of a small speaker and sits right on his desk. Inside is a full Nvidia GPU running AI workloads 24/7. He rents out the compute power to companies that need local AI processing. They pay him a flat monthly fee to use it remotely. $8,000 lands in his account every single month. He just plugged it in and that was it. The $4,000 box paid for itself in two weeks. Now it prints $96,000 a year from a device smaller than a coffee machine. Companies are desperate for local compute and regular people are quietly supplying it. Save this before everyone figures out what he already knows.

winkle.

57,314 views • 15 days ago

NEW: Microsoft just unveiled its most powerful Surface ever. Surface Laptop Ultra is a 15-inch NVIDIA-powered laptop built to run local AI models, creative work, and gaming. • NVIDIA RTX Spark chip with up to 20 Arm CPU cores • Blackwell GPU with 6,144 CUDA cores • Up to 128GB unified memory • Runs up to 120B parameter AI models locally • 15-inch mini-LED display with 2,000 nits peak brightness • Dual-fan cooling with 2.5x the thermal capacity of Surface Laptop 7 This could be a serious AI workstation.

NEW: Microsoft just unveiled its most powerful Surface ever. Surface Laptop Ultra is a 15-inch NVIDIA-powered laptop built to run local AI models, creative work, and gaming. • NVIDIA RTX Spark chip with up to 20 Arm CPU cores • Blackwell GPU with 6,144 CUDA cores • Up to 128GB unified memory • Runs up to 120B parameter AI models locally • 15-inch mini-LED display with 2,000 nits peak brightness • Dual-fan cooling with 2.5x the thermal capacity of Surface Laptop 7 This could be a serious AI workstation.

Ritwik Pavan

93,146 views • 13 days ago

CHINESE AI TEACHER STACKED 4 NVIDIA DGX SPARK CHIPS INTO A CLUSTER - AND NOW RUNS A BUSINESS ON IT FROM HIS DESK 4 chips at $2,999 each - one time investment - and now he has a cluster that handles client inference requests around the clock while he does something else he's running a Qwen 32 35B model with 4 concurrent requests at the same time and the speed barely drops compared to a single request what used to require a $10,000+/month cloud setup now sits on his desk and costs him $40-60/month in electricity 4 chips at $2,999 each - one time investment - and now he has a cluster that handles client inference requests around the clock while he does something else the people still paying per token are financing someone else's infrastructure

CHINESE AI TEACHER STACKED 4 NVIDIA DGX SPARK CHIPS INTO A CLUSTER - AND NOW RUNS A BUSINESS ON IT FROM HIS DESK 4 chips at $2,999 each - one time investment - and now he has a cluster that handles client inference requests around the clock while he does something else he's running a Qwen 32 35B model with 4 concurrent requests at the same time and the speed barely drops compared to a single request what used to require a $10,000+/month cloud setup now sits on his desk and costs him $40-60/month in electricity 4 chips at $2,999 each - one time investment - and now he has a cluster that handles client inference requests around the clock while he does something else the people still paying per token are financing someone else's infrastructure

Sprytix

50,912 views • 15 days ago

Qwen 3.6 35B running at over 100 tokens per second on my $5,399 MacBook Pro M5 Max. This is the best local AI model I have ever run. 128GB of unified memory. No cloud. No API costs. No rate limits. Just raw local inference at speeds I didn't think were possible on a laptop. This model is more intelligent than GPT 5 on benchmarks. Running locally. On a MacBook. For free after the hardware cost. I said local AI would never compete with frontier. I'm starting to rethink that. The gap is closing faster than anyone expected.

Qwen 3.6 35B running at over 100 tokens per second on my $5,399 MacBook Pro M5 Max. This is the best local AI model I have ever run. 128GB of unified memory. No cloud. No API costs. No rate limits. Just raw local inference at speeds I didn't think were possible on a laptop. This model is more intelligent than GPT 5 on benchmarks. Running locally. On a MacBook. For free after the hardware cost. I said local AI would never compete with frontier. I'm starting to rethink that. The gap is closing faster than anyone expected.

BridgeMind

61,978 views • 19 days ago

NVIDIA sent us 2 DGX Sparks. For a while we wondered what we would do with them. The memory bandwidth is 273GB/s making it 3x slower than an M3 Ultra (819GB/s) for batch_size=1 inference. But it has 4x more FLOPS (100 TFLOPS compared to 26 TFLOPS). So we thought, what if we could combine the DGX Spark & M3 Ultra, and make use of both the massive compute on the DGX Spark and the massive memory-bandwidth on the M3 Ultra. We came up with a way to split inference across both devices and achieve a speedup of up to 4x for long prompts compared to the M3 Ultra on its own. Full details in the blog post linked below.

NVIDIA sent us 2 DGX Sparks. For a while we wondered what we would do with them. The memory bandwidth is 273GB/s making it 3x slower than an M3 Ultra (819GB/s) for batch_size=1 inference. But it has 4x more FLOPS (100 TFLOPS compared to 26 TFLOPS). So we thought, what if we could combine the DGX Spark & M3 Ultra, and make use of both the massive compute on the DGX Spark and the massive memory-bandwidth on the M3 Ultra. We came up with a way to split inference across both devices and achieve a speedup of up to 4x for long prompts compared to the M3 Ultra on its own. Full details in the blog post linked below.

Alex Cheema

281,225 views • 8 months ago

Gemma 4 31B is way too slow on my MacBook Pro M5 Max with 128GB of unified memory. 42 seconds to generate the first line of code. 21 tokens per second after that. Uses 26GB of memory when loaded. I sat there waiting nearly a minute before it even started writing code. The output? Actually decent. The lava lamp it generated looks good. But 42 seconds to first token is a dealbreaker. Claude Opus 4.7 starts generating in under 2 seconds. GPT 5.5 is instant. Good model. Unusable speed. Local inference on a $5,399 machine still can't match the cloud.

Gemma 4 31B is way too slow on my MacBook Pro M5 Max with 128GB of unified memory. 42 seconds to generate the first line of code. 21 tokens per second after that. Uses 26GB of memory when loaded. I sat there waiting nearly a minute before it even started writing code. The output? Actually decent. The lava lamp it generated looks good. But 42 seconds to first token is a dealbreaker. Claude Opus 4.7 starts generating in under 2 seconds. GPT 5.5 is instant. Good model. Unusable speed. Local inference on a $5,399 machine still can't match the cloud.

BridgeMind

39,870 views • 17 days ago

NVIDIA will cut your power bill if you let them bolt a $1,000,000 data center to your house. Or spend $2,999 and make $22,000. Option 1, they bolt 16 Blackwell chips to the outside of your house. You get free internet, battery backup, cheaper electricity. They get a million-dollar asset running 24/7 on your property, serving requests for Amazon, Microsoft, Google. Option 2, you buy a DGX Spark for $2,999. Size of a paperback. Sits on your desk. 128GB memory. Runs a 70B model locally. $10/month in electricity. No data leaves the room. Paid for itself in 6 weeks for someone spending $1,900/month on cloud GPU rentals. $22,000 back in year one. One option, you're the infrastructure. Other option, you own it. Same NVIDIA chips. Very different contracts.

NVIDIA will cut your power bill if you let them bolt a $1,000,000 data center to your house. Or spend $2,999 and make $22,000. Option 1, they bolt 16 Blackwell chips to the outside of your house. You get free internet, battery backup, cheaper electricity. They get a million-dollar asset running 24/7 on your property, serving requests for Amazon, Microsoft, Google. Option 2, you buy a DGX Spark for $2,999. Size of a paperback. Sits on your desk. 128GB memory. Runs a 70B model locally. $10/month in electricity. No data leaves the room. Paid for itself in 6 weeks for someone spending $1,900/month on cloud GPU rentals. $22,000 back in year one. One option, you're the infrastructure. Other option, you own it. Same NVIDIA chips. Very different contracts.

Superior

386,382 views • 16 days ago

💫 Introducing our new NVIDIA DGX Spark which belongs to a new class of computers designed to build and run #AI, enabling you to prototype, fine-tune and #inference large AI models locally and seamlessly deploy to the data center or cloud. #SparkSomethingBig #GTC25

💫 Introducing our new NVIDIA DGX Spark which belongs to a new class of computers designed to build and run #AI, enabling you to prototype, fine-tune and #inference large AI models locally and seamlessly deploy to the data center or cloud. #SparkSomethingBig #GTC25

NVIDIA AI Developer

14,480 views • 1 year ago