Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Whew! Big work today getting optimized llama.cpp MIPS R8000 kernel running on the SGI Power Challenge deskside from 1995 with Gemma 3 270M. Went from 0.16 to 0.5 tok/s with a combination of MIPSPro compiler and manual asm optimizations. Not bad for 30 year old hardware. 1/

mov_axbx's profile picture

Nathan Odle

6,874 subscribers

38,079 Aufrufe • vor 1 Monat •via X (Twitter)

KunstWissenschaft & Technologie
LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Let me demonstrate the true power of llama.cpp: - Running on Mac Studio M2 Ultra (3 years old) - Gemma 4 26B A4B Q8_0 (full quality) - Built-in WebUI (ships with llama.cpp) - MCP support out of the box (web-search, HF, github, etc.) - Prompt speculative decoding The result: 300t/s (realtime video)
0:59

Let me demonstrate the true power of llama.cpp: - Running on Mac Studio M2 Ultra (3 years old) - Gemma 4 26B A4B Q8_0 (full quality) - Built-in WebUI (ships with llama.cpp) - MCP support out of the box (web-search, HF, github, etc.) - Prompt speculative decoding The result: 300t/s (realtime video)

Georgi Gerganov

783,099 Aufrufe • vor 2 Monaten

Vinix, an OS/kernel written in pure V, booting on real hardware in a second and running the C++ compiler!
0:35

Vinix, an OS/kernel written in pure V, booting on real hardware in a second and running the C++ compiler!

The V Programming Language

39,100 Aufrufe • vor 2 Jahren

Apple Silicon + Gemma 4 fans: this is for you. Pico AI Server now supports continuous batching with MLX-Swift. 43 tok/s on 1 stream. 26 tok/s per stream on 2 concurrent streams. That’s 52 tok/s total. a 21% throughput gain on a six-year-old MacBook Pro M1 Max!
1:41

Apple Silicon + Gemma 4 fans: this is for you. Pico AI Server now supports continuous batching with MLX-Swift. 43 tok/s on 1 stream. 26 tok/s per stream on 2 concurrent streams. That’s 52 tok/s total. a 21% throughput gain on a six-year-old MacBook Pro M1 Max!

Ronald Mannak

48,116 Aufrufe • vor 1 Monat

You can now have an AI researcher running on your laptop 24/7 for free! Running Qwen3-35B-A3B with llama.cpp and a 4-bit quant from Unsloth
0:30

You can now have an AI researcher running on your laptop 24/7 for free! Running Qwen3-35B-A3B with llama.cpp and a 4-bit quant from Unsloth

Lewis Tunstall

117,829 Aufrufe • vor 1 Monat

I just fine-tuned Google DeepMind's "gemma-3-270m" and it's ready for secure deployment in production! → Imported and fine-tuned using Hugging Face transformers on M4 → Packaged and versioned with KitOps → Pushed to Jozu Hub for production use and deployments All within just a few minutes🔥 "gemma-3-270m" is lightweight, powerful, and designed for production-grade systems
2:56

I just fine-tuned Google DeepMind's "gemma-3-270m" and it's ready for secure deployment in production! → Imported and fine-tuned using Hugging Face transformers on M4 → Packaged and versioned with KitOps → Pushed to Jozu Hub for production use and deployments All within just a few minutes🔥 "gemma-3-270m" is lightweight, powerful, and designed for production-grade systems

Mr. Ånand

53,804 Aufrufe • vor 9 Monaten

Introducing AMD Ryzen AI Halo, an Agent Computer built to make local AI development fast and simple. • Ryzen AI Developer Center pre-installed • Pre-optimized from first power on • Built to reduce setup friction • Designed for local AI development AMD Ryzen AI Halo brings AI from the cloud into the hands of developers with Ryzen AI developer center pre-installed and optimized from first power on, developers can hit the ground running.
1:07

Introducing AMD Ryzen AI Halo, an Agent Computer built to make local AI development fast and simple. • Ryzen AI Developer Center pre-installed • Pre-optimized from first power on • Built to reduce setup friction • Designed for local AI development AMD Ryzen AI Halo brings AI from the cloud into the hands of developers with Ryzen AI developer center pre-installed and optimized from first power on, developers can hit the ground running.

AMD

25,704 Aufrufe • vor 16 Tagen

BruceDropEmoff goes off on YourRage and Bendadon for being 30 and claiming to have a platonic relationship with 19-year-old Yonna “It’s not okay for a 30 year old to have a platonic relationships with a 19 year old. What do they have in common ? ”
0:51

BruceDropEmoff goes off on YourRage and Bendadon for being 30 and claiming to have a platonic relationship with 19-year-old Yonna “It’s not okay for a 30 year old to have a platonic relationships with a 19 year old. What do they have in common ? ”

validate

1,616,762 Aufrufe • vor 1 Monat

i pointed hermes agent at nvidia's nemotron cascade 2 30B-A3B on a single RTX 3090 24GB. IQ4_XS quant by bartowski, 187 tok/s, 625K context. had it discover its own hardware, create an identity file, then build a full GPU marketplace UI from a single prompt. it one shotted it. first attempt no iteration. qwen 3.5 35B-A3B on the same hardware same 3090 24GB took an iteration to recover from a blank screen on the same type of build. 24 days between these two models releasing. same active parameters, completely different architectures and cascade 2 through hermes agent just keeps going. this model goes on and on. feast your eyes. more iterations and tests dropping soon. nvidia really cooked. no special flags needed. nvidia optimized this mamba MoE so well it just runs. flash attention auto enabled, context auto allocated. the model does the work not the config. but i compiled llama.cpp from source and i'm not sure how it performs on other engines. if you ran nemotron on any hardware drop your numbers below. RTX, AMD, Mac, whatever. model, quant, tok/s, engine. i want to see if it holds everywhere or just on llama.cpp.
1:31

i pointed hermes agent at nvidia's nemotron cascade 2 30B-A3B on a single RTX 3090 24GB. IQ4_XS quant by bartowski, 187 tok/s, 625K context. had it discover its own hardware, create an identity file, then build a full GPU marketplace UI from a single prompt. it one shotted it. first attempt no iteration. qwen 3.5 35B-A3B on the same hardware same 3090 24GB took an iteration to recover from a blank screen on the same type of build. 24 days between these two models releasing. same active parameters, completely different architectures and cascade 2 through hermes agent just keeps going. this model goes on and on. feast your eyes. more iterations and tests dropping soon. nvidia really cooked. no special flags needed. nvidia optimized this mamba MoE so well it just runs. flash attention auto enabled, context auto allocated. the model does the work not the config. but i compiled llama.cpp from source and i'm not sure how it performs on other engines. if you ran nemotron on any hardware drop your numbers below. RTX, AMD, Mac, whatever. model, quant, tok/s, engine. i want to see if it holds everywhere or just on llama.cpp.

Sudo su

70,583 Aufrufe • vor 2 Monaten

llama.cpp with MTP support makes local models fast enough to use as daily drivers 🚀 Qwen3.6-27B dense generation (on A10G): From 25 tok/s → 45 tok/s (+78%). Two flags on llama-server: --spec-type draft-mtp --spec-draft-n-max 2
0:38

llama.cpp with MTP support makes local models fast enough to use as daily drivers 🚀 Qwen3.6-27B dense generation (on A10G): From 25 tok/s → 45 tok/s (+78%). Two flags on llama-server: --spec-type draft-mtp --spec-draft-n-max 2

Victor M

171,172 Aufrufe • vor 25 Tagen

1/ Autoresearch from Andrej Karpathy has been one of the most interesting agentic patterns to emerge this year. The challenge: right now every agent runs experiments in isolation, duplicating work and compute, forgetting findings, rediscovering dead ends. Everyone is running in solo mode! Today I'm releasing Community Computer: a collaborative network for autoresearch-like code experiments. 💻 Agents conduct experiments, publish signed results, and build on each other's work. The community reproduces findings on their own hardware. 👇
1:36

1/ Autoresearch from Andrej Karpathy has been one of the most interesting agentic patterns to emerge this year. The challenge: right now every agent runs experiments in isolation, duplicating work and compute, forgetting findings, rediscovering dead ends. Everyone is running in solo mode! Today I'm releasing Community Computer: a collaborative network for autoresearch-like code experiments. 💻 Agents conduct experiments, publish signed results, and build on each other's work. The community reproduces findings on their own hardware. 👇

Eleftherios

15,762 Aufrufe • vor 1 Monat

Footage of a 3 year old chimney sweep from the 1930's.
0:30

Footage of a 3 year old chimney sweep from the 1930's.

Historic Vids

4,361,824 Aufrufe • vor 3 Jahren

"Making Minimal Techno with WebAssembly And a C Compiler" My presentation from #JSHeroes2024 today 👉
2:03

"Making Minimal Techno with WebAssembly And a C Compiler" My presentation from #JSHeroes2024 today 👉

Tero Parviainen

15,554 Aufrufe • vor 2 Jahren

Back-to-back days with a bad challenge from Ben Rice and the Yankees are out of challenges He walked on the next pitch
0:23

Back-to-back days with a bad challenge from Ben Rice and the Yankees are out of challenges He walked on the next pitch

Talkin' Yanks

107,611 Aufrufe • vor 2 Monaten

* You are filled with the power of learning not to snap on beat 1 and 3.
1:58

Sensitive content

* You are filled with the power of learning not to snap on beat 1 and 3.

SQUIFF

144,469 Aufrufe • vor 11 Monaten

A 25-year-old Muslim man from Bihar went to work in Kangra, Himachal Pradesh and locals gave him a room on rent. Just days later, he abducted an 8-year-old Hindu girl from the neighboring house and took her into the jungle. Her parents with police help, caught him at midnight.
3:23

A 25-year-old Muslim man from Bihar went to work in Kangra, Himachal Pradesh and locals gave him a room on rent. Just days later, he abducted an 8-year-old Hindu girl from the neighboring house and took her into the jungle. Her parents with police help, caught him at midnight.

Team Hindu United

86,217 Aufrufe • vor 1 Monat

Last week we went out into the wilderness, far from any cell service or Wi-Fi, to test an experimental Gemma 4 app, all 100% offline: 🔍Visual understanding 🧠Math reasoning 🛠️Tool use and more Running seamlessly on Pixel hardware and paired with our prototype display glasses.
1:01

Last week we went out into the wilderness, far from any cell service or Wi-Fi, to test an experimental Gemma 4 app, all 100% offline: 🔍Visual understanding 🧠Math reasoning 🛠️Tool use and more Running seamlessly on Pixel hardware and paired with our prototype display glasses.

Google Gemma

103,462 Aufrufe • vor 21 Tagen

Open-source VLA from Galaxea Dynamics shown on a wide range of environments and with different objects, and all running on their cheap off the shelf hardware. Really great contribution to the community.
0:31

Open-source VLA from Galaxea Dynamics shown on a wide range of environments and with different objects, and all running on their cheap off the shelf hardware. Really great contribution to the community.

Chris Paxton

30,125 Aufrufe • vor 5 Monaten

Meet Tobenna, Nigeria's three-year-old football prodigy. Three-year-old Tobenna Uchendu stole hearts in 2023 when a video of him skillfully running with a football at just 1 year and 9 months old went viral.
1:23

Meet Tobenna, Nigeria's three-year-old football prodigy. Three-year-old Tobenna Uchendu stole hearts in 2023 when a video of him skillfully running with a football at just 1 year and 9 months old went viral.

BBC News Africa

279,071 Aufrufe • vor 1 Jahr

Florida got a big spark and much needed contribution from second year guard Urban Klavzar Klavzar produced 15 points, 3 rebounds, 3 assists off the bench and also pitched in 4 3's on 4/7 shooting from deep Hit a huge 3 vs FSU to win that game, playing with a lot of confidence as of late, after playing with zero confidence to begin the year. Needs to be more consistent but can be a nice piece for the Gators this year.
0:55

Florida got a big spark and much needed contribution from second year guard Urban Klavzar Klavzar produced 15 points, 3 rebounds, 3 assists off the bench and also pitched in 4 3's on 4/7 shooting from deep Hit a huge 3 vs FSU to win that game, playing with a lot of confidence as of late, after playing with zero confidence to begin the year. Needs to be more consistent but can be a nice piece for the Gators this year.

Arman Jovic

95,639 Aufrufe • vor 6 Monaten

PLAY OF THE YEAR? PLAY OF THE YEAR! After a challenge and great camera work from BRC13 Sports, Evan Roche comes down with a RIDICOULOUS catch! #SCTop10 x @ESPNassigndesk x #WhereWarriorsBelong
3:53

PLAY OF THE YEAR? PLAY OF THE YEAR! After a challenge and great camera work from BRC13 Sports, Evan Roche comes down with a RIDICOULOUS catch! #SCTop10 x @ESPNassigndesk x #WhereWarriorsBelong

ESU Warriors

24,497 Aufrufe • vor 1 Jahr