Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

~ 6.5 - 6.7 t/s for GLM 5.1 on M5 Max 128GB Added “Dense” model export, now model load is only 5s ! Experts are streaming from SSD, so we do not pre-load it. Added direct SSD->Slot memory path, removed prefetch... Many dead end experiments. See Export a “dense-only...

anemll's profile picture

Anemll

4,014 subscribers

29,421 Aufrufe • vor 4 Monaten •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Initial DeepSeek v4 and v4 Flash support added to anemll-flash-llama.cpp! 🚀 M5 MAX 128GB can run full DeepSeek-V4 with 1.6T params in original FP8/FP4 weights from SSD without requantization! One-click scripts to convert HF safetensors to dense + MoE sidecar (no GGUF needed) Inference and server examples Prefill and decoding benchmarks Branch and docs here: This is WIP and I plan to add cleanup and more testing this week. Feedback welcome!
0:21

Initial DeepSeek v4 and v4 Flash support added to anemll-flash-llama.cpp! 🚀 M5 MAX 128GB can run full DeepSeek-V4 with 1.6T params in original FP8/FP4 weights from SSD without requantization! One-click scripts to convert HF safetensors to dense + MoE sidecar (no GGUF needed) Inference and server examples Prefill and decoding benchmarks Branch and docs here: This is WIP and I plan to add cleanup and more testing this week. Feedback welcome!

Anemll

32,250 Aufrufe • vor 3 Monaten

WIP: First attempt to speed up prefill for Flash-MoE. Original repo did token-by-token without streamed experts. Added: Batched linear attention + batched full attention (Flash Attention style) with custom Metal kernels. Without experts: 6.2x faster prefill (11 -> 68 tok/s) With experts at full-attn layers only: 1.9x faster (11 -> 20.5 tok/s) — same output quality Qwen3.5-397B, 4-bit, 209GB, M5 Max 128GB 1/3
0:57

WIP: First attempt to speed up prefill for Flash-MoE. Original repo did token-by-token without streamed experts. Added: Batched linear attention + batched full attention (Flash Attention style) with custom Metal kernels. Without experts: 6.2x faster prefill (11 -> 68 tok/s) With experts at full-attn layers only: 1.9x faster (11 -> 20.5 tok/s) — same output quality Qwen3.5-397B, 4-bit, 209GB, M5 Max 128GB 1/3

Anemll

19,572 Aufrufe • vor 5 Monaten

Apple’s “LLM in a Flash” is definitely worth checking out. Going to 2-bit for the shared-expert MLP means disk I/O is no longer dominant. 14–15 tok/s from SSD is still wild for a ~400B MoE model streamed from storage. Qwen3.5-397B-A17B Credit: Dan Woods
0:23

Apple’s “LLM in a Flash” is definitely worth checking out. Going to 2-bit for the shared-expert MLP means disk I/O is no longer dominant. 14–15 tok/s from SSD is still wild for a ~400B MoE model streamed from storage. Qwen3.5-397B-A17B Credit: Dan Woods

Anemll

35,664 Aufrufe • vor 5 Monaten

Added few tweaks to alexintosh iOS port. ~7 t/s on iPhone 17pro Also tested on Neo, looks like A19pro SSD is way faster! The same prompt…
1:12

Added few tweaks to alexintosh iOS port. ~7 t/s on iPhone 17pro Also tested on Neo, looks like A19pro SSD is way faster! The same prompt…

Anemll

40,461 Aufrufe • vor 5 Monaten

I’ve been asked if external SSD works ? Here is M4 Pro 24GB running MinMax 2.7 @ 7.7 fps Unsloth AI quant IQ2_XXS @ 73GB MOE_TOPK=4, --moe-slot-bank 48, It's using ACASIS USB4v2 80 Gbps enclosure with “budget” T710 1TB Gen5 SSD over TB5 connection. I’m also testing different enclosure and SSDs. It seems QD1/Q1T1 random access is the most critical. Llama.cpp experimental fork/branch with MinMax-2.7: Quants and benchmarks are here: ACASIS Official
2:51

I’ve been asked if external SSD works ? Here is M4 Pro 24GB running MinMax 2.7 @ 7.7 fps Unsloth AI quant IQ2_XXS @ 73GB MOE_TOPK=4, --moe-slot-bank 48, It's using ACASIS USB4v2 80 Gbps enclosure with “budget” T710 1TB Gen5 SSD over TB5 connection. I’m also testing different enclosure and SSDs. It seems QD1/Q1T1 random access is the most critical. Llama.cpp experimental fork/branch with MinMax-2.7: Quants and benchmarks are here: ACASIS Official

Anemll

18,597 Aufrufe • vor 4 Monaten

This is what 30 hours of work, a handful of AI bux, and and a sore back looks like: 2x RTX 6000 Pro 320GB RAM (192V+128GB) 48,128 CUDA Cores 1.8TB/s Memory Bandwidth 9950X (16c/32t) Corsair DDR5 6400 ASUS ProArt X870E 4TB Crucial T705 5x4 SSD 4TB Crucial P3 4x4 SSD 4TB Crucial T500 4x4 SSD Asrock Platinum 1600W Corsair XD5 Reservoir EKWB Waterblock 2x 360mm EKWB Radiators Way too much debt I'm so stupid But it's glorious I bought everything used or repurposed old equipment except for the water block, 1 GPU, and 3 fans. Even though I'm not thrilled about spending the cash, I only have about $10.5K into it (not counting parts I paid for years ago). Safe to say I did pretty damn good all things considered. I'd like to thank Central Computers for helping me get this project off the ground. Now that I have this together, my aim is to support the OSS community. If you're interested in running local models, consider following me on this journey. Next chapter loading....
0:26

This is what 30 hours of work, a handful of AI bux, and and a sore back looks like: 2x RTX 6000 Pro 320GB RAM (192V+128GB) 48,128 CUDA Cores 1.8TB/s Memory Bandwidth 9950X (16c/32t) Corsair DDR5 6400 ASUS ProArt X870E 4TB Crucial T705 5x4 SSD 4TB Crucial P3 4x4 SSD 4TB Crucial T500 4x4 SSD Asrock Platinum 1600W Corsair XD5 Reservoir EKWB Waterblock 2x 360mm EKWB Radiators Way too much debt I'm so stupid But it's glorious I bought everything used or repurposed old equipment except for the water block, 1 GPU, and 3 fans. Even though I'm not thrilled about spending the cash, I only have about $10.5K into it (not counting parts I paid for years ago). Safe to say I did pretty damn good all things considered. I'd like to thank Central Computers for helping me get this project off the ground. Now that I have this together, my aim is to support the OSS community. If you're interested in running local models, consider following me on this journey. Next chapter loading....

am.will

48,629 Aufrufe • vor 4 Monaten

New mactop w/ANE readout for M5 of macOS 27 Confirmed! Just upgraded M5 MAX, it need something different to get ANE :( Looking into it If you have M3U upgraded to 27 please test Ivan Fioravanti ᯅ Simple command like this triggers ANE: fm respond --model system --stream 'make a game of Space invaders in PyGame'
0:21

New mactop w/ANE readout for M5 of macOS 27 Confirmed! Just upgraded M5 MAX, it need something different to get ANE :( Looking into it If you have M3U upgraded to 27 please test Ivan Fioravanti ᯅ Simple command like this triggers ANE: fm respond --model system --stream 'make a game of Space invaders in PyGame'

Anemll

12,637 Aufrufe • vor 2 Monaten

This is the fastest I've seen Llama 3.3 running anywhere! Llama 3.3 70B running at 652 t/s is lightning fast. And if you want Llama 3.1, here are the speeds I was able to get: • Llama 3.1 8B: 1006 t/s • Llama 3.1 70B: 709 t/s • Llama 3.1 405B: 206 t/s (You can access all of these models for free! See the link below.) This speed is incredible, but the interesting part is what's happening behind the scenes: These models aren't running on a GPU! In this video, I'm using the SambaNova cloud to access these models. They built a custom chip (SN40L) optimized for AI workflows. A single SN40L chip can hold hundreds of models (trillions of parameters) in memory! The speed alone is a huge deal, but the big advantage is for agentic workflows running multiple specialized models. A GPU can only host a single model and switch (unload and load) to a different model if necessary. An SN40L, on the other hand, can host every model at once, making it much faster. Here is the video where you can see how fast these chips are:
7:24

This is the fastest I've seen Llama 3.3 running anywhere! Llama 3.3 70B running at 652 t/s is lightning fast. And if you want Llama 3.1, here are the speeds I was able to get: • Llama 3.1 8B: 1006 t/s • Llama 3.1 70B: 709 t/s • Llama 3.1 405B: 206 t/s (You can access all of these models for free! See the link below.) This speed is incredible, but the interesting part is what's happening behind the scenes: These models aren't running on a GPU! In this video, I'm using the SambaNova cloud to access these models. They built a custom chip (SN40L) optimized for AI workflows. A single SN40L chip can hold hundreds of models (trillions of parameters) in memory! The speed alone is a huge deal, but the big advantage is for agentic workflows running multiple specialized models. A GPU can only host a single model and switch (unload and load) to a different model if necessary. An SN40L, on the other hand, can host every model at once, making it much faster. Here is the video where you can see how fast these chips are:

Santiago

97,543 Aufrufe • vor 1 Jahr

So many little UX/UI edits with the change from web to mobile! So many people said 'just do a wrapper, or pwa'. So glad we went native... coming in a few weeks! We've... - ⏱️ moved the List and End time info from the bottom, all the way to the top. Because they are there to be 'seen' but not 'tapped' - 🟡 added a FAB button to make it easier to add tasks when your list is long - 🧭 added tab bar navigation - woo!! - added modals instead of drop-downs for interactions - 📳 added haptics! and they seriously make it so awesome - 🐣 included some easter eggs - Lulu is now on the screen and tapping her will reveal Barry.. and other stuff (later)
0:22

So many little UX/UI edits with the change from web to mobile! So many people said 'just do a wrapper, or pwa'. So glad we went native... coming in a few weeks! We've... - ⏱️ moved the List and End time info from the bottom, all the way to the top. Because they are there to be 'seen' but not 'tapped' - 🟡 added a FAB button to make it easier to add tasks when your list is long - 🧭 added tab bar navigation - woo!! - added modals instead of drop-downs for interactions - 📳 added haptics! and they seriously make it so awesome - 🐣 included some easter eggs - Lulu is now on the screen and tapping her will reveal Barry.. and other stuff (later)

Marie Ng 🦙

10,329 Aufrufe • vor 2 Jahren

the tiebreaker is done. qwen 3.5 27B dense. single RTX 3090. one prompt. zero steering. zero human edits. 1,827 lines across 10 files. 13 minutes. full thinking mode. runs on first load. hermes 4.3 got the same prompt with 2x 3090s and 5x the context it needed. wrote 1,249 lines, left empty files, needed 3 interventions, game was broken on load. same architecture class. same quant. hermes got double the hardware. completely different result. dense wasn't the problem. hermes was. but here's what got me. this model thinks at 27 tok/s. every single token carries 27 billion parameters of reasoning. MoE hit 112 tok/s but only 3B active per token. the dense model is slower and it doesn't matter. watch 13 minutes of autonomous coding on a consumer GPU with zero intervention and tell me speed is what matters. a year ago this wasn't possible. now it runs on hardware you can buy used for $900. no API. no subscription. no cloud. just a 3090 doing what data centers did 18 months ago. full unedited session in the video. every token, every file, every thinking chain. 16 minutes. hit play.
16:04

the tiebreaker is done. qwen 3.5 27B dense. single RTX 3090. one prompt. zero steering. zero human edits. 1,827 lines across 10 files. 13 minutes. full thinking mode. runs on first load. hermes 4.3 got the same prompt with 2x 3090s and 5x the context it needed. wrote 1,249 lines, left empty files, needed 3 interventions, game was broken on load. same architecture class. same quant. hermes got double the hardware. completely different result. dense wasn't the problem. hermes was. but here's what got me. this model thinks at 27 tok/s. every single token carries 27 billion parameters of reasoning. MoE hit 112 tok/s but only 3B active per token. the dense model is slower and it doesn't matter. watch 13 minutes of autonomous coding on a consumer GPU with zero intervention and tell me speed is what matters. a year ago this wasn't possible. now it runs on hardware you can buy used for $900. no API. no subscription. no cloud. just a 3090 doing what data centers did 18 months ago. full unedited session in the video. every token, every file, every thinking chain. 16 minutes. hit play.

Sudo su

91,135 Aufrufe • vor 5 Monaten

Running 400B model on iPhone! 0.6 t/s Credit Dan Woods @alexintosh Daniel Isaac Anemll
3:51

Running 400B model on iPhone! 0.6 t/s Credit Dan Woods @alexintosh Daniel Isaac Anemll

Anemll

294,484 Aufrufe • vor 5 Monaten

Here is gemma-4-26B-A4B-it on A17 Pro chip w/8GB memory ( MacBook Neo) ~ 7 t/s running on AMX ( GPU is slower on A17) Gemma's 4 expert is x2.3 larger than Qwen See Qwen 35B below
0:41

Here is gemma-4-26B-A4B-it on A17 Pro chip w/8GB memory ( MacBook Neo) ~ 7 t/s running on AMX ( GPU is slower on A17) Gemma's 4 expert is x2.3 larger than Qwen See Qwen 35B below

Anemll

195,282 Aufrufe • vor 4 Monaten

DELL ASUS ACER AND GIGABYTE ALL SHIPPED $2,999 DGX SPARK CLONES IN THE SAME QUARTER, SAMSUNG GEN 5 SSD CHOICE CAUSES 2X PERFORMANCE GAPS, THE LOCAL AI HARDWARE MAP JUST GREW BY FOUR FRONTIER TIER OPTIONS 02:13 the reviewer flips through the spec sheets, "we saw major differences in the SSD, some are using Gen 4 SSDs which will slow these platforms down, others have Gen 5 Samsung drives, Samsung is the leader by almost a factor of two" NVIDIA's GB10 Grace Blackwell reference board got picked up by Dell, Asus, Acer and Gigabyte in the same quarter, four $2,999 mini boxes with 128GB unified memory and 1 PFLOP of AI throughput from competing OEMs the Asus version runs coolest across GPU, CPU and SSD thanks to an unpainted bottom plate and a copper heatsink element, the Acer leads on price-to-perf, the Dell honeycomb intake actually restricts airflow slightly, the gigabyte sits middle of pack Samsung Gen 5 versus Gen 4 inside the same chassis causes a 2x read-write gap for the exact same $2,999 sticker, the buyer who reads the SSD spec sheet gets a 70B model loading in 4 seconds, the one who doesn't waits 8 your serie's frontier tier was the $1,700 EVO-X2, four DGX Spark OEM clones at $2,999 extend the map upward, the local AI hardware boom is now staffed by every major PC vendor not just AMD and apple, the category fully arrived bookmark this and read the article below
2:50

DELL ASUS ACER AND GIGABYTE ALL SHIPPED $2,999 DGX SPARK CLONES IN THE SAME QUARTER, SAMSUNG GEN 5 SSD CHOICE CAUSES 2X PERFORMANCE GAPS, THE LOCAL AI HARDWARE MAP JUST GREW BY FOUR FRONTIER TIER OPTIONS 02:13 the reviewer flips through the spec sheets, "we saw major differences in the SSD, some are using Gen 4 SSDs which will slow these platforms down, others have Gen 5 Samsung drives, Samsung is the leader by almost a factor of two" NVIDIA's GB10 Grace Blackwell reference board got picked up by Dell, Asus, Acer and Gigabyte in the same quarter, four $2,999 mini boxes with 128GB unified memory and 1 PFLOP of AI throughput from competing OEMs the Asus version runs coolest across GPU, CPU and SSD thanks to an unpainted bottom plate and a copper heatsink element, the Acer leads on price-to-perf, the Dell honeycomb intake actually restricts airflow slightly, the gigabyte sits middle of pack Samsung Gen 5 versus Gen 4 inside the same chassis causes a 2x read-write gap for the exact same $2,999 sticker, the buyer who reads the SSD spec sheet gets a 70B model loading in 4 seconds, the one who doesn't waits 8 your serie's frontier tier was the $1,700 EVO-X2, four DGX Spark OEM clones at $2,999 extend the map upward, the local AI hardware boom is now staffed by every major PC vendor not just AMD and apple, the category fully arrived bookmark this and read the article below

starmex

17,573 Aufrufe • vor 1 Monat

Alibaba just released a coding model that hits 82 percent on SWE-Bench Verified. That is the highest score ever published for an open-source model. The weights are free. The license is Apache 2.0. You can run it today. The model is Qwen 4 Coder 32B. Here is what 82 percent on SWE-Bench Verified actually means. SWE-Bench Verified tests whether an AI can autonomously resolve real bugs pulled from real production GitHub repositories. Not synthetic exercises. Real open-source projects that real teams depend on. A model gets a bug report, reads the code, writes a fix, and either passes the test suite or it does not. At 82 percent, Qwen 4 Coder 32B resolves 82 out of every 100 real production bugs it is given. Without a human guiding it. On code it has never seen before. For comparison: Qwen 4 Coder 32B: 82 percent SWE-Bench Verified. Open source. Apache 2.0. Claude Fable 5: 80.3 percent SWE-Bench Pro. $10 input / $50 output per million tokens. Currently suspended. GPT-5.6 Sol: Competitive on Terminal-Bench. $5 input / $30 output per million tokens. An open-weight model that you can download and run for free just beat both of them on the benchmark designed to measure real software engineering capability. Here is the architecture. Qwen 4 Coder 32B is a 32 billion parameter dense model. Not a Mixture-of-Experts. Every parameter is active on every request. This matters for inference: a dense 32B model runs on 22 gigabytes of VRAM, which fits on a single high-end consumer GPU or a MacBook Pro with 64GB of unified memory. The smaller variant, Qwen 4 Coder 4B, runs at approximately 135 tokens per second on an M5 Max and fits inside 8 gigabytes of RAM. For a model with usable coding capability, that is a new bar for what fits in a single laptop. The training methodology continued Alibaba's approach of reinforcement learning on verifiable coding tasks. The model gets rewarded when its code passes tests. It gets penalized when it fails. Over millions of training steps, the model learns to write code that actually runs rather than code that looks plausible. License: Apache 2.0. Full commercial use. No attribution requirement. No revenue threshold. No monthly active user ceiling. Weights: Hugging Face, available today. Runs on: vLLM, Ollama, SGLang, and any standard GGUF-compatible inference engine. Qwen 4 32B also runs at approximately 135 tokens per second on an M5 Max chip, setting a new bar for what a sub-8GB model can do on Apple Silicon. The open-source coding model just beat the best closed-source model in the world on the benchmark designed to test whether AI can actually do software engineering. The weights are free. The subscription is optional. Source: Autom8Labs AI Insight July 2026, State of Open Source LLMs June 2026, Kunal Ganglani blog June 2026.
0:25

Alibaba just released a coding model that hits 82 percent on SWE-Bench Verified. That is the highest score ever published for an open-source model. The weights are free. The license is Apache 2.0. You can run it today. The model is Qwen 4 Coder 32B. Here is what 82 percent on SWE-Bench Verified actually means. SWE-Bench Verified tests whether an AI can autonomously resolve real bugs pulled from real production GitHub repositories. Not synthetic exercises. Real open-source projects that real teams depend on. A model gets a bug report, reads the code, writes a fix, and either passes the test suite or it does not. At 82 percent, Qwen 4 Coder 32B resolves 82 out of every 100 real production bugs it is given. Without a human guiding it. On code it has never seen before. For comparison: Qwen 4 Coder 32B: 82 percent SWE-Bench Verified. Open source. Apache 2.0. Claude Fable 5: 80.3 percent SWE-Bench Pro. $10 input / $50 output per million tokens. Currently suspended. GPT-5.6 Sol: Competitive on Terminal-Bench. $5 input / $30 output per million tokens. An open-weight model that you can download and run for free just beat both of them on the benchmark designed to measure real software engineering capability. Here is the architecture. Qwen 4 Coder 32B is a 32 billion parameter dense model. Not a Mixture-of-Experts. Every parameter is active on every request. This matters for inference: a dense 32B model runs on 22 gigabytes of VRAM, which fits on a single high-end consumer GPU or a MacBook Pro with 64GB of unified memory. The smaller variant, Qwen 4 Coder 4B, runs at approximately 135 tokens per second on an M5 Max and fits inside 8 gigabytes of RAM. For a model with usable coding capability, that is a new bar for what fits in a single laptop. The training methodology continued Alibaba's approach of reinforcement learning on verifiable coding tasks. The model gets rewarded when its code passes tests. It gets penalized when it fails. Over millions of training steps, the model learns to write code that actually runs rather than code that looks plausible. License: Apache 2.0. Full commercial use. No attribution requirement. No revenue threshold. No monthly active user ceiling. Weights: Hugging Face, available today. Runs on: vLLM, Ollama, SGLang, and any standard GGUF-compatible inference engine. Qwen 4 32B also runs at approximately 135 tokens per second on an M5 Max chip, setting a new bar for what a sub-8GB model can do on Apple Silicon. The open-source coding model just beat the best closed-source model in the world on the benchmark designed to test whether AI can actually do software engineering. The weights are free. The subscription is optional. Source: Autom8Labs AI Insight July 2026, State of Open Source LLMs June 2026, Kunal Ganglani blog June 2026.

Harman

41,278 Aufrufe • vor 1 Monat

First impressions on Muse Glimmer! It's incredibly fast for a dense model, currently running an average of 208tps with a max of 274tps on a single 5090 with their DFLASH config. Comparatively, though, both using Open Code, Qwopus Coder (with thinking off) produced a much better shark survival game than the one I got from Glimmer. Meta's new dense model is currently just lacking some HTML canvas taste, but this is something that can be added via SFT as long as the model is stable and capable from a back-end programming perspective. And it seems to be, without a doubt. The big kicker here is that I ran this at extra high thinking, and it did not take long at all to run. Our current local leader, Qwen 27B 3.6, has a tendency to overthink, but with glimmer, that is not the case. Right now, my recommendation for general local programming (Apps, Games, Websites, Visual Tools) in this class is still Qwopus Coder with thinking disabled, or Qwopus Fusion with thinking enabled. Of course Shark Survival is a very basic domain-specific test, but I find that the result scales very well across many domains. If we're going to be shipping apps generated entirely locally, visual taste is somewhat of a bare minimum requirement, solely in my opinion, and Qwen's models in this class offer significantly more at the moment. That's actually why I initially started getting into finetuning with Qwen 3.5, they were the first base that was able to do really good front-end with some opus-trace fine-tuning. Qwen 3.6 has taste even in the base model, and we know Qwen 3.8 is going to blow us all away! Regardless, this looks like a very tempting new base model. As a first offering from Meta in this class for a long time, I am incredibly impressed and elated to have it. We now finally have a proper Single GPU frontier race, instead of us just begging Qwen for more releases. Single GPU open frontier model race is a VERY good thing. Please keep pushing Meta
1:20

First impressions on Muse Glimmer! It's incredibly fast for a dense model, currently running an average of 208tps with a max of 274tps on a single 5090 with their DFLASH config. Comparatively, though, both using Open Code, Qwopus Coder (with thinking off) produced a much better shark survival game than the one I got from Glimmer. Meta's new dense model is currently just lacking some HTML canvas taste, but this is something that can be added via SFT as long as the model is stable and capable from a back-end programming perspective. And it seems to be, without a doubt. The big kicker here is that I ran this at extra high thinking, and it did not take long at all to run. Our current local leader, Qwen 27B 3.6, has a tendency to overthink, but with glimmer, that is not the case. Right now, my recommendation for general local programming (Apps, Games, Websites, Visual Tools) in this class is still Qwopus Coder with thinking disabled, or Qwopus Fusion with thinking enabled. Of course Shark Survival is a very basic domain-specific test, but I find that the result scales very well across many domains. If we're going to be shipping apps generated entirely locally, visual taste is somewhat of a bare minimum requirement, solely in my opinion, and Qwen's models in this class offer significantly more at the moment. That's actually why I initially started getting into finetuning with Qwen 3.5, they were the first base that was able to do really good front-end with some opus-trace fine-tuning. Qwen 3.6 has taste even in the base model, and we know Qwen 3.8 is going to blow us all away! Regardless, this looks like a very tempting new base model. As a first offering from Meta in this class for a long time, I am incredibly impressed and elated to have it. We now finally have a proper Single GPU frontier race, instead of us just begging Qwen for more releases. Single GPU open frontier model race is a VERY good thing. Please keep pushing Meta

Kyle Hessling

17,399 Aufrufe • vor 13 Tagen

We talk about the Minnesota youth hockey model a lot, and so I get asked about it a lot. Why does Minnesota have not only the most players, but why do they develop the most elite players as well? One of the main reasons is that the path is simple to understand for players and families. You get into the game playing for your town, you play for your town all the way into high school, and then you go to junior/college hockey if you’re good enough. In so many other places, the path is confusing. Do I play town or club? When do I make the jump to AAA? Do I stay with the organization in my city or do I go to another club? What tier of hockey is right for my kid? The amount of options and decisions often leads to FOMO, confusion, decision fatigue, and a burnout for players and families. It also leads to a lot of jersey chasing and organization hopping rather than playing for something bigger than you, which is huge for a development of a love for the game. If we can simplify the path, tiers can work together rather than against each other, and create a model where families can better understand the path to help their kids achieve their dreams, our sport will grow and we will develop more elite players. We talked all about this on our latest podcast on what we’ve learned about youth hockey.
0:50

We talk about the Minnesota youth hockey model a lot, and so I get asked about it a lot. Why does Minnesota have not only the most players, but why do they develop the most elite players as well? One of the main reasons is that the path is simple to understand for players and families. You get into the game playing for your town, you play for your town all the way into high school, and then you go to junior/college hockey if you’re good enough. In so many other places, the path is confusing. Do I play town or club? When do I make the jump to AAA? Do I stay with the organization in my city or do I go to another club? What tier of hockey is right for my kid? The amount of options and decisions often leads to FOMO, confusion, decision fatigue, and a burnout for players and families. It also leads to a lot of jersey chasing and organization hopping rather than playing for something bigger than you, which is huge for a development of a love for the game. If we can simplify the path, tiers can work together rather than against each other, and create a model where families can better understand the path to help their kids achieve their dreams, our sport will grow and we will develop more elite players. We talked all about this on our latest podcast on what we’ve learned about youth hockey.

Topher Scott

116,938 Aufrufe • vor 24 Tagen

Responsible AI should not feel like the path of most resistance. That was the strongest idea from my SAS Innovate conversation with Reggie Townsend, leading Data Ethics, Governance, and Social Impact at SAS Software. Reggie framed governance not as a compliance layer, but as a way to scale human judgment. Too often, we innovate first and govern later. Model selected. Agent deployed. Process built. Then governance arrives at the end and feels like friction. Bolt it on after the fact, and resistance is guaranteed. The opportunity: design responsible AI, so it becomes intuitive, action-oriented, and useful in the flow of work. That is what Reggie meant by making responsibility "irresistible." Second point: Use cases must lead. When everyone can access the same models, differentiation will not come from the technology. It will come from how leaders define outcomes, govern applications, and connect business value to institutional trust. The risk does not live in the model. The value does not live in the model. Both live in the use case. For CEOs and boards, this is the shift: from model-first oversight to outcome-first accountability. Better questions before scaling AI: - What human decision are we shaping? - What business outcome are we improving? - What risk are we containing? - What judgment are we extending? Responsible AI becomes strategic when it helps people make better decisions, faster, with greater confidence. Most leaders can't see where governance sits inside their AI operating model. SAS AI Navigator makes it visible: Design for the human. Not only the technology.
5:12

Responsible AI should not feel like the path of most resistance. That was the strongest idea from my SAS Innovate conversation with Reggie Townsend, leading Data Ethics, Governance, and Social Impact at SAS Software. Reggie framed governance not as a compliance layer, but as a way to scale human judgment. Too often, we innovate first and govern later. Model selected. Agent deployed. Process built. Then governance arrives at the end and feels like friction. Bolt it on after the fact, and resistance is guaranteed. The opportunity: design responsible AI, so it becomes intuitive, action-oriented, and useful in the flow of work. That is what Reggie meant by making responsibility "irresistible." Second point: Use cases must lead. When everyone can access the same models, differentiation will not come from the technology. It will come from how leaders define outcomes, govern applications, and connect business value to institutional trust. The risk does not live in the model. The value does not live in the model. Both live in the use case. For CEOs and boards, this is the shift: from model-first oversight to outcome-first accountability. Better questions before scaling AI: - What human decision are we shaping? - What business outcome are we improving? - What risk are we containing? - What judgment are we extending? Responsible AI becomes strategic when it helps people make better decisions, faster, with greater confidence. Most leaders can't see where governance sits inside their AI operating model. SAS AI Navigator makes it visible: Design for the human. Not only the technology.

Sabine VanderLinden

835,838 Aufrufe • vor 3 Monaten

Here is the first look of the Financial Model for Pakistan Stock Exchange we have built. At this moment, I don't know if it will be successful or not. Yet it's backtesting has given promising results. I believe it is real time and future testing that matters the most. To make it transparent, I will keep on updating on "X" the top picks from the Model. I request you not to invest a single money based on it's results because it is in the early stage. Every feedback from your side would be highly encouraged. The problem was that investors only see MARI, FFC, ,OGDC PPL, MEBL, Luck like companies because they are mainstream. But remember that there are 457 listed companies and there are some hidden gems that are usually missed by retail investors like us. One such example was SSOM that just exploded in few weeks. Thats why, the purpose of the model was to find those companies that have a lot of potential but are not covered by any brokerage house and yet there results are excellent. Based on this model, some time back, I added positions in companies like AGIL (Agri Autos) SPEL (Synthetic Papers ltd), NATF (National Foods), KSB Pumps, Treet. Because scores of all these companies is excellent. They remained stagnant for long time so my conviction on the model was also declining. But it were last two weeks that made a difference and so was my own conviction on this model. Big names like Wealth ⚡️ Wise Value Investor Doctor in PSX Tayyab 🇵🇰 are already aware of it. On Thursday I added HCAR. I will keep you updated just for testing purpose I again repeat not to invest. I will risk my own money. If it is successful, we intend to make a website where you all have access to this data. How it works and how to read it, I will explain it to you in a separate thread.
0:52

Here is the first look of the Financial Model for Pakistan Stock Exchange we have built. At this moment, I don't know if it will be successful or not. Yet it's backtesting has given promising results. I believe it is real time and future testing that matters the most. To make it transparent, I will keep on updating on "X" the top picks from the Model. I request you not to invest a single money based on it's results because it is in the early stage. Every feedback from your side would be highly encouraged. The problem was that investors only see MARI, FFC, ,OGDC PPL, MEBL, Luck like companies because they are mainstream. But remember that there are 457 listed companies and there are some hidden gems that are usually missed by retail investors like us. One such example was SSOM that just exploded in few weeks. Thats why, the purpose of the model was to find those companies that have a lot of potential but are not covered by any brokerage house and yet there results are excellent. Based on this model, some time back, I added positions in companies like AGIL (Agri Autos) SPEL (Synthetic Papers ltd), NATF (National Foods), KSB Pumps, Treet. Because scores of all these companies is excellent. They remained stagnant for long time so my conviction on the model was also declining. But it were last two weeks that made a difference and so was my own conviction on this model. Big names like Wealth ⚡️ Wise Value Investor Doctor in PSX Tayyab 🇵🇰 are already aware of it. On Thursday I added HCAR. I will keep you updated just for testing purpose I again repeat not to invest. I will risk my own money. If it is successful, we intend to make a website where you all have access to this data. How it works and how to read it, I will explain it to you in a separate thread.

Shayan Baig

54,727 Aufrufe • vor 1 Jahr

My Latest AI short was made with Seedance 2.0 the model that has been firing up the internet over the past couple of days. I made this in two and half days and it's an original short that is based on HP Lovecraft's work which is now Public Domain (No Copyright infringement here!). If you are not into period romances then bare with it as it gets a more little action packed in the second half! The model is brilliant at both performance and action, you can now make convincing dialogue and action sequences for AI cinema. With this model we have now reached the stage where we can tell any story we want with AI, the tools are capable. This is only 720p at the moment but I suspect it will be upgraded to 1080p very soon. We stand at the threshold of having broadcast quality tools capable of creating anything we can imagine available to everyone, so lets do new an interesting things with it. Thank you to CapCut for whitelisting me and to Dreamina AI AI for giving me access as part of their CPP. Thanks for watching! 👀
3:29

My Latest AI short was made with Seedance 2.0 the model that has been firing up the internet over the past couple of days. I made this in two and half days and it's an original short that is based on HP Lovecraft's work which is now Public Domain (No Copyright infringement here!). If you are not into period romances then bare with it as it gets a more little action packed in the second half! The model is brilliant at both performance and action, you can now make convincing dialogue and action sequences for AI cinema. With this model we have now reached the stage where we can tell any story we want with AI, the tools are capable. This is only 720p at the moment but I suspect it will be upgraded to 1080p very soon. We stand at the threshold of having broadcast quality tools capable of creating anything we can imagine available to everyone, so lets do new an interesting things with it. Thank you to CapCut for whitelisting me and to Dreamina AI AI for giving me access as part of their CPP. Thanks for watching! 👀

Uncanny Harry AI

76,824 Aufrufe • vor 6 Monaten