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.

Drew Timme vs Sioux Falls: 36PTS-16REB-8AST-16/27FG

Laker Performances

19,264 subscribers

61,493 views • 5 months ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Drew Timme vs Blazers: - 21 Points - 4 Assists - 87.5 TS%

Drew Timme vs Blazers: - 21 Points - 4 Assists - 87.5 TS%

Laker Performances

244,066 views • 8 months ago

Drew Timme tonight vs OKC: ▪️22 PTS ▪️9 REB ▪️10-13 FG

Drew Timme tonight vs OKC: ▪️22 PTS ▪️9 REB ▪️10-13 FG

Brooklyn Nets

83,777 views • 1 year ago

Bronny James vs Sioux Falls: 26PTS-7REB-5AST-1STL-9/12FG-6/8 3PT

Bronny James vs Sioux Falls: 26PTS-7REB-5AST-1STL-9/12FG-6/8 3PT

Laker Performances

339,114 views • 5 months ago

vs. SIOUX FALLS 36 MIN | 26 PTS | 7 REB | 5 AST | 75% FG | 6/8 3FG 🔥

vs. SIOUX FALLS 36 MIN | 26 PTS | 7 REB | 5 AST | 75% FG | 6/8 3FG 🔥

Bronny Tribute

10,685 views • 5 months ago

If someone's mean to you, act stupid. IMPULSE BUY TOUR: SEPT 10: FARGO, ND SEPT 11: SIOUX FALLS, SD (SOLD OUT) SEPT 12: CEDAR RAPIDS, IA SEPT 15-16: GRAND RAPIDS, MI SEPT 18: BOISE, ID (LOW TIX) SEPT 19: COEUR D'ALENE, ID (LOW TIX) #airports #comedyreels #flights #flightdelays

If someone's mean to you, act stupid. IMPULSE BUY TOUR: SEPT 10: FARGO, ND SEPT 11: SIOUX FALLS, SD (SOLD OUT) SEPT 12: CEDAR RAPIDS, IA SEPT 15-16: GRAND RAPIDS, MI SEPT 18: BOISE, ID (LOW TIX) SEPT 19: COEUR D'ALENE, ID (LOW TIX) #airports #comedyreels #flights #flightdelays

Greg Warren

52,520 views • 9 days ago

Unsigned senior Trey Parker receives his first D2 offer from Sioux Falls following an exceptional performance at this year’s state tournament with Apple Valley. State Quarterfinal: 17 pts vs No. 2 Wayzata State Semifinal: 23 pts vs No. 3 Maple Grove State Championship: 33 pts vs No. 5 Chaska Scored a team-high in each game, shooting 48% from the field, 44% from three, and 85% from the line.

Unsigned senior Trey Parker receives his first D2 offer from Sioux Falls following an exceptional performance at this year’s state tournament with Apple Valley. State Quarterfinal: 17 pts vs No. 2 Wayzata State Semifinal: 23 pts vs No. 3 Maple Grove State Championship: 33 pts vs No. 5 Chaska Scored a team-high in each game, shooting 48% from the field, 44% from three, and 85% from the line.

And-One Films

25,257 views • 5 months ago

Jamaree Bouyea was UNSTOPPABLE during his rookie season! The Sioux Falls Skyforce star averaged 18.6 PPG, 6.4 APG, and 5.1 RPG, which earned him All-Rookie and All-G League Second Team honors. 🙌 Bouyea was one of four G League players to average 16+ PPG, 6+ APG, and 5+ RPG.

Jamaree Bouyea was UNSTOPPABLE during his rookie season! The Sioux Falls Skyforce star averaged 18.6 PPG, 6.4 APG, and 5.1 RPG, which earned him All-Rookie and All-G League Second Team honors. 🙌 Bouyea was one of four G League players to average 16+ PPG, 6+ APG, and 5+ RPG.

NBA G League

47,315 views • 3 years ago

“I don’t think this [loss] falls on Pochettino.” Taylor Twellman on who’s most at fault for the USMNT’s Round of 16 loss vs. Belgium. (via Yahoo Sports Daily)

“I don’t think this [loss] falls on Pochettino.” Taylor Twellman on who’s most at fault for the USMNT’s Round of 16 loss vs. Belgium. (via Yahoo Sports Daily)

Yahoo Sports

25,355 views • 2 months ago

MLB Hitters To Watch (5/23) Trent Grisham - .273 AVG & .818 OPS with 1 HR in 11 AB vs. Drew Rasmussen Taylor Ward - .484 AVG & 1.095 OPS in 31 AB vs. Framber Valdez Adley Rutschman - .375 AVG & 1.037 OPS with 1 HR in 16 AB vs. Framber Valdez Pete Alonso - .400 AVG & .900 OPS in 10 AB vs. Framber Valdez Jarren Duran - .308 AVG & .938 OPS in 13 AB vs. Taj Bradley Mini-thread🧵

MLB Hitters To Watch (5/23) Trent Grisham - .273 AVG & .818 OPS with 1 HR in 11 AB vs. Drew Rasmussen Taylor Ward - .484 AVG & 1.095 OPS in 31 AB vs. Framber Valdez Adley Rutschman - .375 AVG & 1.037 OPS with 1 HR in 16 AB vs. Framber Valdez Pete Alonso - .400 AVG & .900 OPS in 10 AB vs. Framber Valdez Jarren Duran - .308 AVG & .938 OPS in 13 AB vs. Taj Bradley Mini-thread🧵

Tobey Schulman

10,507 views • 3 months ago

⚾️9/5 MLB MOST Notables⚾️ Show support for what we do👉 Wilyer Abreu vs Chris Bassitt: 8-16, 2 Doubles, 2 HRs💣💣 Salvador Perez vs Max Scherzer: 14-38, 2 Doubles, 4 HRs💣💣💣💣 Carter Jensen vs Max Scherzer: 3-4, 2 Doubles, 106.6 EV Seiya Suzuki vs Ryan Gusto: 1-2, 1 HR💣 Dominic Smith vs Zack Wheeler: 13-33, 1 Double, 1 HR💣 Ronald Acuña Jr. vs Zack Wheeler: 13-56, 4 Doubles, 4 HRs💣💣💣💣, 16 Ks🤔 Trea Turner vs Martín Pérez: 5-17, 2 Doubles, 1 HR💣 Alec Bohm vs Martín Pérez: 8-20, 1 HR💣 Bryan De La Cruz vs Martín Pérez: 4-6, 1 Double Riley Greene vs Parker Messick: 2-4, 1 HR💣, 102.8 EV Eduardo Valencia vs Parker Messick: 1-3, 1 HR💣 José Ramírez vs Framber Valdez: 6-16, 2 Doubles, 1 HR💣 Jo Adell vs Framber Valdez: 7-25, 2 Doubles, 1 Triple, 2 HRs💣💣 Yusei Kikuchi vs Current Pirates: 5-32, 9 K🔥 William Contreras vs Andrew Abbott: 6-20, 1 Double, 1 HR💣 Jackson Chourio vs Andrew Abbott: 3-8, 2 HRs💣💣 Brice Turang vs Andrew Abbott: 4-10, 3 Doubles Garrett Mitchell vs Andrew Abbott: 3-3, 2 Doubles, 103.8 EV Eugenio Suárez vs Dustin May: 4-9, 1 HR💣 Eli White vs Chris Bassitt: 5-11, 2 Doubles Adley Rutschman vs Chris Bassitt: 5-16, 3 Doubles Pete Alonso vs Sonny Gray: 4-11, 2 HRs💣💣 Gunnar Henderson vs Sonny Gray: 2-6, 1 HR💣 Luis Robert Jr. vs Sonny Gray: 4-12, 1 Double, 1 HR💣 Jonathan Aranda vs Jacob deGrom: 2-3, 1 Double, 1 HR💣, 105.8 EV Jake Burger vs Drew Rasmussen: 3-8, 1 HR💣 Brandon Nimmo vs Drew Rasmussen: 3-3 Andrés Giménez vs Seth Lugo: 3-9, 2 Doubles Brooks Lee vs Anthony Kay: 1-2, 1 HR💣 Kyle Teel vs Taj Bradley: 5-7, 1 Double Colson Montgomery vs Taj Bradley: 2-8, 2 HRs💣💣 LaMonte Wade Jr. vs Brandon Pfaadt: 3-10, 1 Double, 1 HR💣 Yainer Diaz vs Brandon Pfaadt: 3-5, 2 Doubles Yordan Alvarez vs Brandon Pfaadt: 1-1, 1 HR💣 Amed Rosario vs Robbie Ray: 6-17, 4 Doubles Xander Bogaerts vs Carlos Rodón: 6-14 Hunter Goodman vs Matthew Liberatore: 2-5, 1 HR💣 Jake McCarthy vs Matthew Liberatore: 2-3, 1 HR💣 Cole Carrigg vs Matthew Liberatore: 2-3, 2 Doubles CJ Abrams vs Tyler Glasnow: 2-4, 1 Double, 1 HR💣 Zack Gelof vs George Kirby: 4-7, 2 Doubles, 1 HR💣 Randy Arozarena vs Jeffrey Springs: 4-14, 1 HR💣 🍓Fresh Matchups (never faced): Zac Thornton vs SFG, Ethan Pecko vs ARI, Mason Adams vs STL Who do YOU🫵 think adds to their notable history today/tonight?

⚾️9/5 MLB MOST Notables⚾️ Show support for what we do👉 Wilyer Abreu vs Chris Bassitt: 8-16, 2 Doubles, 2 HRs💣💣 Salvador Perez vs Max Scherzer: 14-38, 2 Doubles, 4 HRs💣💣💣💣 Carter Jensen vs Max Scherzer: 3-4, 2 Doubles, 106.6 EV Seiya Suzuki vs Ryan Gusto: 1-2, 1 HR💣 Dominic Smith vs Zack Wheeler: 13-33, 1 Double, 1 HR💣 Ronald Acuña Jr. vs Zack Wheeler: 13-56, 4 Doubles, 4 HRs💣💣💣💣, 16 Ks🤔 Trea Turner vs Martín Pérez: 5-17, 2 Doubles, 1 HR💣 Alec Bohm vs Martín Pérez: 8-20, 1 HR💣 Bryan De La Cruz vs Martín Pérez: 4-6, 1 Double Riley Greene vs Parker Messick: 2-4, 1 HR💣, 102.8 EV Eduardo Valencia vs Parker Messick: 1-3, 1 HR💣 José Ramírez vs Framber Valdez: 6-16, 2 Doubles, 1 HR💣 Jo Adell vs Framber Valdez: 7-25, 2 Doubles, 1 Triple, 2 HRs💣💣 Yusei Kikuchi vs Current Pirates: 5-32, 9 K🔥 William Contreras vs Andrew Abbott: 6-20, 1 Double, 1 HR💣 Jackson Chourio vs Andrew Abbott: 3-8, 2 HRs💣💣 Brice Turang vs Andrew Abbott: 4-10, 3 Doubles Garrett Mitchell vs Andrew Abbott: 3-3, 2 Doubles, 103.8 EV Eugenio Suárez vs Dustin May: 4-9, 1 HR💣 Eli White vs Chris Bassitt: 5-11, 2 Doubles Adley Rutschman vs Chris Bassitt: 5-16, 3 Doubles Pete Alonso vs Sonny Gray: 4-11, 2 HRs💣💣 Gunnar Henderson vs Sonny Gray: 2-6, 1 HR💣 Luis Robert Jr. vs Sonny Gray: 4-12, 1 Double, 1 HR💣 Jonathan Aranda vs Jacob deGrom: 2-3, 1 Double, 1 HR💣, 105.8 EV Jake Burger vs Drew Rasmussen: 3-8, 1 HR💣 Brandon Nimmo vs Drew Rasmussen: 3-3 Andrés Giménez vs Seth Lugo: 3-9, 2 Doubles Brooks Lee vs Anthony Kay: 1-2, 1 HR💣 Kyle Teel vs Taj Bradley: 5-7, 1 Double Colson Montgomery vs Taj Bradley: 2-8, 2 HRs💣💣 LaMonte Wade Jr. vs Brandon Pfaadt: 3-10, 1 Double, 1 HR💣 Yainer Diaz vs Brandon Pfaadt: 3-5, 2 Doubles Yordan Alvarez vs Brandon Pfaadt: 1-1, 1 HR💣 Amed Rosario vs Robbie Ray: 6-17, 4 Doubles Xander Bogaerts vs Carlos Rodón: 6-14 Hunter Goodman vs Matthew Liberatore: 2-5, 1 HR💣 Jake McCarthy vs Matthew Liberatore: 2-3, 1 HR💣 Cole Carrigg vs Matthew Liberatore: 2-3, 2 Doubles CJ Abrams vs Tyler Glasnow: 2-4, 1 Double, 1 HR💣 Zack Gelof vs George Kirby: 4-7, 2 Doubles, 1 HR💣 Randy Arozarena vs Jeffrey Springs: 4-14, 1 HR💣 🍓Fresh Matchups (never faced): Zac Thornton vs SFG, Ethan Pecko vs ARI, Mason Adams vs STL Who do YOU🫵 think adds to their notable history today/tonight?

Tablesetters: A Baseball Podcast

45,403 views • 12 days ago

Bits + Bips: Reasons to Be Optimistic After Bitcoin Falls Toward $100K Crypto panic 😱 vs. institutional patience 💼 On Bits + Bips, Bitwise’s Teddy Fusaro joins Ram Ahluwalia CFA, Lumida & Christopher Perkins 🦅🌎⚓️NYC to unpack: 📉 Bitcoin’s crash below $100K 🪙 Why Ripple’s momentum can’t be ignored 💸 Tether’s trillion-dollar question 🏛️ CLARITY Act: what are the odds? Timestamps: 🎬 0:00 Intro 📈 3:03 Why Ram is still bullish on markets despite recent carnage 💧 6:03 Why a slingshot economy will soon send liquidity back into risk assets 🏦 9:18 How institutional sentiment now differs sharply from crypto Twitter 🚀 16:43 Inside Bitwise’s clever move to launch $BSOL 🌊 21:20 Why Chris says “you can’t sleep on #Ripple” 🔗 25:28 How TradFi and crypto rails are starting to merge 💳 29:38 Why every company suddenly wants to own the payments layer 💣 30:53 Exploring #Tether’s biggest challenge —and whether it’s worth $500B 📊 39:52 #ETFs vs. digital asset treasuries (#DATs): Teddy’s take ⚖️ 46:48 Why the CLARITY Act is unlikely to pass in 2025

Bits + Bips: Reasons to Be Optimistic After Bitcoin Falls Toward $100K Crypto panic 😱 vs. institutional patience 💼 On Bits + Bips, Bitwise’s Teddy Fusaro joins Ram Ahluwalia CFA, Lumida & Christopher Perkins 🦅🌎⚓️NYC to unpack: 📉 Bitcoin’s crash below $100K 🪙 Why Ripple’s momentum can’t be ignored 💸 Tether’s trillion-dollar question 🏛️ CLARITY Act: what are the odds? Timestamps: 🎬 0:00 Intro 📈 3:03 Why Ram is still bullish on markets despite recent carnage 💧 6:03 Why a slingshot economy will soon send liquidity back into risk assets 🏦 9:18 How institutional sentiment now differs sharply from crypto Twitter 🚀 16:43 Inside Bitwise’s clever move to launch $BSOL 🌊 21:20 Why Chris says “you can’t sleep on #Ripple” 🔗 25:28 How TradFi and crypto rails are starting to merge 💳 29:38 Why every company suddenly wants to own the payments layer 💣 30:53 Exploring #Tether’s biggest challenge —and whether it’s worth $500B 📊 39:52 #ETFs vs. digital asset treasuries (#DATs): Teddy’s take ⚖️ 46:48 Why the CLARITY Act is unlikely to pass in 2025

Laura Shin

29,303 views • 10 months ago

The new Huberman Lab episode is out: How to Improve Your Memory & Cognitive Function at Any Age | Dr. Alan Castel 0:00 Dr. Alan Castel 2:41 What Is Memory?, Reconstruction & Metacognition 4:49 Mnemonics, Remembering Names & Deeper Learning 8:22 The Penny & Apple Logo, Noticing vs Seeing, Learning Through Mistakes 10:43 Sponsors: Wealthfront & Helix 14:05 Neuroplasticity, Frustration, Curiosity & Mindset 17:42 Maintaining vs Learning New Things, Habits, Novelty & Emotional Memory 24:28 "Mental Photographs," Photo-Taking & Imagining the Future 29:28 Eyewitness Memory, the Ronald Cotton Case, Confidence vs Accuracy 35:07 Medium-Term & Prospective Memory, Hotel Fire Exits 40:28 Sponsor: AG1 41:47 When Habits Turn Lethal, Aviation & Human Error 49:01 Why Memory Changes With Age; Alzheimer's & the Nun Study 52:34 Exercise & Hippocampal Volume, Falls & Balance 57:14 SuperAgers & Athletes; Regret, Balance & Being Driven 1:12:08 Sponsor: Function 1:13:45 Age Stereotypes, Subjective Age & Positive Age Beliefs 1:20:02 Goals & Plans, Scams; Anterior Midcingulate Cortex & SuperAgers 1:26:23 Culture, Resilience, Blue Zones & COVID 1:29:18 Adversity, the Positivity Effect & Intergenerational Learning 1:36:31 Sponsor: Lingo 1:38:00 Limitations & Purpose; Time, Family & Connection 1:44:58 Deliberately Building Memories; the ABCs of Successful Aging 1:51:02 Following Your Interests; Castel's Path & Older Adults 1:57:16 Mental Simulations, Curiosity Studies & Selectivity 2:01:19 Socioemotional Selectivity Theory; Steve Jobs & Lifespan 2:07:10 The Secret to Successful Aging; State vs Trait Curiosity 2:11:04 Scams & AI Voice Cloning 2:14:31 John Wooden, Wisdom, Love & Balance 2:17:41 Learning Through Mistakes; Does the Brain Get Better With Age? 2:25:00 Conclusion, Better With Age 2:26:00 Zero-Cost Support, YouTube, Spotify & Apple Follow, Reviews & Feedback, Sponsors, Protocols Book, Social Media, Neural Network Newsletter Includes paid partnerships.

Andrew D. Huberman, Ph.D.

127,501 views • 2 months ago

these UGC creators have driven 5 billion+ views for consumer apps in the past 90 days. I talked to Drew Levin from SideShift and he broke down the viral formats working right now (it's not just hook & demo anymore) full vid👇 0:00 - Sourcing UGC from apps driving 500M+ views 0:15 - Engineering virality and leveraging AI controversy 3:59 - The shift from massive influencer campaigns to authentic UGC 4:38 - New format strategy: 10-minute long-form story videos 5:57 - Using rage bait and health trends to capture watch time 8:40 - Scaling UGC programs with campaign managers and content coaches 9:29 - Creator Managers (logistics) vs. Content Coaches (creative feedback) 10:41 - The winning manager-to-creator scaling ratio 13:00 - Finding golden formats and frictionless creator onboarding 16:40 - Format Deconstruction: A 9.6M-view talking head case study 18:21 - Using engagement rate as a predictive virality indicator 24:44 - The "Duolingo-style" training program to mint new talent 29:00 - Case Study: Reaching #1 on the App Store via Spark Ads 29:44 - Protecting budgets: Isolating paid spend from creator bonuses 32:28 - Algorithmic and UI differences: Meta Partnership vs. TikTok Spark Ads 34:21 - The infinite growth glitch: Traffic quality of organic-boosted spend 35:45 - The 3-5x rule: Using paid spend to lift organic shadowbans 42:38 - Creator compensation: Retainer base vs. stacked milestone structures 44:06 - Payout optimization: Capitalizing on the "free view" framework 45:45 - What to expect: Realistic CPM progression over the first 90 days

these UGC creators have driven 5 billion+ views for consumer apps in the past 90 days. I talked to Drew Levin from SideShift and he broke down the viral formats working right now (it's not just hook & demo anymore) full vid👇 0:00 - Sourcing UGC from apps driving 500M+ views 0:15 - Engineering virality and leveraging AI controversy 3:59 - The shift from massive influencer campaigns to authentic UGC 4:38 - New format strategy: 10-minute long-form story videos 5:57 - Using rage bait and health trends to capture watch time 8:40 - Scaling UGC programs with campaign managers and content coaches 9:29 - Creator Managers (logistics) vs. Content Coaches (creative feedback) 10:41 - The winning manager-to-creator scaling ratio 13:00 - Finding golden formats and frictionless creator onboarding 16:40 - Format Deconstruction: A 9.6M-view talking head case study 18:21 - Using engagement rate as a predictive virality indicator 24:44 - The "Duolingo-style" training program to mint new talent 29:00 - Case Study: Reaching #1 on the App Store via Spark Ads 29:44 - Protecting budgets: Isolating paid spend from creator bonuses 32:28 - Algorithmic and UI differences: Meta Partnership vs. TikTok Spark Ads 34:21 - The infinite growth glitch: Traffic quality of organic-boosted spend 35:45 - The 3-5x rule: Using paid spend to lift organic shadowbans 42:38 - Creator compensation: Retainer base vs. stacked milestone structures 44:06 - Payout optimization: Capitalizing on the "free view" framework 45:45 - What to expect: Realistic CPM progression over the first 90 days

Joseph Choi

22,595 views • 3 months ago

$PrismML Releases Bonsai 27B: 1-bit and Ternary Builds of Qwen3.6-27B Hitting 89.5% of FP16 at 3.9GB. No new pretrain. No higher-precision escape hatches. No multi-GPU rig. Here's how it works. 👇 1: Codes, not floats Every weight becomes a code, with one shared FP16 scale per group of 128. Ternary is {−1, 0, +1}, binary is {−1, +1}. Sharing the scale across 128 weights keeps its cost at 16/128 = 0.125 bits. → Ternary: log2(3) + 16/128 ≈ 1.71 bits/weight → 5.9GB → Binary: 1 + 16/128 = 1.125 bits/weight → 3.9GB 2: Post-training, not from scratch No BitNet-style low-bit pretrain. It starts from off-the-shelf Qwen3.6-27B, architecture unchanged. The representation runs end to end across embeddings, attention projections, MLP projections, and the LM head. → 9.4× (ternary) and 14.2× (binary) vs the 54GB FP16 baseline 3: Labels are not bit-widths Conventional low-bit builds are mixed-precision by construction. The advertised name describes the most-compressed tensors, not the model. → Q4_K_XL, labeled "4-bit," is really 5.2 bits/weight at 17.6GB → IQ2_XXS, labeled "2-bit," is really 2.8 bits/weight at 9.4GB 4: Fitting a phone is two budgets iOS caps a single app near half of RAM, so a 12GB iPhone exposes ~6GB. The KV cache grows on top. Hybrid attention at ~75% linear means only 16 of 64 layers cache. → 4-bit KV: 4.3GB at 262K context, down from 17.2GB → 11.0 tok/s on iPhone 17 Pro Max 5: The numbers (15 benchmarks, thinking mode) → Ternary: 80.49 avg at 5.9GB — 94.6% of FP16 → 1-bit: 76.11 avg at 3.9GB — 89.5% of FP16 → IQ2_XXS falls to 57.5 on AIME26 while still scoring 88.93 on MMLU-Redux The key takeaway: 27B-class reasoning without the 54GB checkpoint — group-wise ternary and binary codes, an end-to-end low-bit language stack, 4-bit KV, on one phone. Full analysis: Repo: Model weight: Technical details: PrismML$

PrismML Releases Bonsai 27B: 1-bit and Ternary Builds of Qwen3.6-27B Hitting 89.5% of FP16 at 3.9GB. No new pretrain. No higher-precision escape hatches. No multi-GPU rig. Here's how it works. 👇 1: Codes, not floats Every weight becomes a code, with one shared FP16 scale per group of 128. Ternary is {−1, 0, +1}, binary is {−1, +1}. Sharing the scale across 128 weights keeps its cost at 16/128 = 0.125 bits. → Ternary: log2(3) + 16/128 ≈ 1.71 bits/weight → 5.9GB → Binary: 1 + 16/128 = 1.125 bits/weight → 3.9GB 2: Post-training, not from scratch No BitNet-style low-bit pretrain. It starts from off-the-shelf Qwen3.6-27B, architecture unchanged. The representation runs end to end across embeddings, attention projections, MLP projections, and the LM head. → 9.4× (ternary) and 14.2× (binary) vs the 54GB FP16 baseline 3: Labels are not bit-widths Conventional low-bit builds are mixed-precision by construction. The advertised name describes the most-compressed tensors, not the model. → Q4_K_XL, labeled "4-bit," is really 5.2 bits/weight at 17.6GB → IQ2_XXS, labeled "2-bit," is really 2.8 bits/weight at 9.4GB 4: Fitting a phone is two budgets iOS caps a single app near half of RAM, so a 12GB iPhone exposes ~6GB. The KV cache grows on top. Hybrid attention at ~75% linear means only 16 of 64 layers cache. → 4-bit KV: 4.3GB at 262K context, down from 17.2GB → 11.0 tok/s on iPhone 17 Pro Max 5: The numbers (15 benchmarks, thinking mode) → Ternary: 80.49 avg at 5.9GB — 94.6% of FP16 → 1-bit: 76.11 avg at 3.9GB — 89.5% of FP16 → IQ2_XXS falls to 57.5 on AIME26 while still scoring 88.93 on MMLU-Redux The key takeaway: 27B-class reasoning without the 54GB checkpoint — group-wise ternary and binary codes, an end-to-end low-bit language stack, 4-bit KV, on one phone. Full analysis: Repo: Model weight: Technical details: PrismML

Marktechpost AI

31,860 views • 2 months ago

Inside the mind of an ex-SIG quant trader who can't turn off the EV brain - even for his kid's school choice Andrew Courtney (Andrew Courtney) ran the International ETFs Trading Desk at Susquehanna International Group for ~15 years before leaving in 2023. He now runs Kalshionomics (Kalshinomics), a prediction markets analytics tool, and writes the Whirligig Bear, one of the sharpest prediction markets Substacks out there. "I think of everything as a bet. I kind of don't understand how you talk to normal people — they do not do that." SIG trains their junior traders with poker, spending 2hrs/day turning over cards after every hand, justifying every decision quantitatively AND qualitatively. 15 years later, Andrew views prediction markets the same way: read who's on the other side, size accordingly, fold when the whale comes back at you 10x. We cover: - Why SIG pays junior traders to play poker for 2hrs/day — & what happens after every single hand - The "one eye on the market, always" attention tax that destroys most people's careers - How to find edge in prediction markets by asking: who am I actually trading against? - Why meme-heavy, overhyped markets (Taylor Swift at the Super Bowl) might be the juiciest trades - The insider trading debate in prediction markets — & why it's "socially corrosive" - Floor trading vs. upstairs quant: why the transition saved his career - 40 connections after ~15 years at one of the world's best firms — the hidden cost of prop trading - Why he doesn't have collision insurance on his car (& the EV math behind it) Thank you so much Andrew Courtney for coming on the pod! Timestamps: 00:00 Intro 05:00 Floor trading vs. electronic trading 06:28 What makes an upstairs trader 10:16 Poker as trader training 13:00 Thinking in bets as a mental framework 15:11 Decision trees in real life 16:40 Where prediction markets actually have edge 19:00 Why the LLM forecasting layer falls short 19:40 Liquidity incentives and trading low-volume markets 22:00 Limiting downside even when the model is wrong 24:32 Executing in illiquid markets 25:44 Fair value vs. directional conviction 27:11 Bayesian updating when liquidity responds 28:40 Fading hype and crowded narratives 31:07 Longshot bias vs. fanbase bias 34:20 How to judge whether you really have edge 36:40 Building analytics tools for prediction markets 38:20 The temporary edge for smart amateurs 40:35 Where prediction markets fit best 41:20 Markets that shouldn’t exist 43:20 Why insider trading corrodes incentives 46:52 Are prediction markets a net good or bad 50:47 Minimizing degeneracy and maximizing signal 53:32 A simple EV mindset anyone can use

Inside the mind of an ex-SIG quant trader who can't turn off the EV brain - even for his kid's school choice Andrew Courtney (Andrew Courtney) ran the International ETFs Trading Desk at Susquehanna International Group for ~15 years before leaving in 2023. He now runs Kalshionomics (Kalshinomics), a prediction markets analytics tool, and writes the Whirligig Bear, one of the sharpest prediction markets Substacks out there. "I think of everything as a bet. I kind of don't understand how you talk to normal people — they do not do that." SIG trains their junior traders with poker, spending 2hrs/day turning over cards after every hand, justifying every decision quantitatively AND qualitatively. 15 years later, Andrew views prediction markets the same way: read who's on the other side, size accordingly, fold when the whale comes back at you 10x. We cover: - Why SIG pays junior traders to play poker for 2hrs/day — & what happens after every single hand - The "one eye on the market, always" attention tax that destroys most people's careers - How to find edge in prediction markets by asking: who am I actually trading against? - Why meme-heavy, overhyped markets (Taylor Swift at the Super Bowl) might be the juiciest trades - The insider trading debate in prediction markets — & why it's "socially corrosive" - Floor trading vs. upstairs quant: why the transition saved his career - 40 connections after ~15 years at one of the world's best firms — the hidden cost of prop trading - Why he doesn't have collision insurance on his car (& the EV math behind it) Thank you so much Andrew Courtney for coming on the pod! Timestamps: 00:00 Intro 05:00 Floor trading vs. electronic trading 06:28 What makes an upstairs trader 10:16 Poker as trader training 13:00 Thinking in bets as a mental framework 15:11 Decision trees in real life 16:40 Where prediction markets actually have edge 19:00 Why the LLM forecasting layer falls short 19:40 Liquidity incentives and trading low-volume markets 22:00 Limiting downside even when the model is wrong 24:32 Executing in illiquid markets 25:44 Fair value vs. directional conviction 27:11 Bayesian updating when liquidity responds 28:40 Fading hype and crowded narratives 31:07 Longshot bias vs. fanbase bias 34:20 How to judge whether you really have edge 36:40 Building analytics tools for prediction markets 38:20 The temporary edge for smart amateurs 40:35 Where prediction markets fit best 41:20 Markets that shouldn’t exist 43:20 Why insider trading corrodes incentives 46:52 Are prediction markets a net good or bad 50:47 Minimizing degeneracy and maximizing signal 53:32 A simple EV mindset anyone can use

Ethan Kho

436,074 views • 7 months ago