Загрузка видео...

Не удалось загрузить видео

Возникла проблема при загрузке этого видео. Это может быть связано с временными проблемами сети или видео может быть недоступно.

На главную

4/ to achieve maximum memory efficiency, we quantize model weights to ~4-bit, getting the language model under 20GB with room for the kv cache, perception encoder, and drafter alongside it. a dflash drafter proposes blocks of tokens the main model verifies in parallel, so it stays responsive.

Alexandr Wang

627,327 subscribers

67,151 просмотров • 13 дней назад •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

We took a 30B model and split it in two to write tokens in parallel instead of one at a time. Introducing Nemotron-Labs-TwoTower: a diffusion language model from NVIDIA Research adapted from Nemotron-3-Nano-30B-A3B. Here’s how it works: one half holds the context, the other writes the tokens, with both reusing the pretrained model instead of training a new one from scratch. We found it kept 98.7% of the original model’s quality at 2.42× faster generation.

We took a 30B model and split it in two to write tokens in parallel instead of one at a time. Introducing Nemotron-Labs-TwoTower: a diffusion language model from NVIDIA Research adapted from Nemotron-3-Nano-30B-A3B. Here’s how it works: one half holds the context, the other writes the tokens, with both reusing the pretrained model instead of training a new one from scratch. We found it kept 98.7% of the original model’s quality at 2.42× faster generation.

NVIDIA AI

761,317 просмотров • 1 месяц назад

We’re releasing model weights for our 8B- parameter Dynamic Byte Latent Transformer, an alternative to traditional tokenization methods with the potential to redefine the standards for language model efficiency and reliability. Learn more about how Dynamic Byte Latent Transformer is paving the way for groundbreaking developments in the field of language modeling. Read the research paper, and download the model and code.

We’re releasing model weights for our 8B- parameter Dynamic Byte Latent Transformer, an alternative to traditional tokenization methods with the potential to redefine the standards for language model efficiency and reliability. Learn more about how Dynamic Byte Latent Transformer is paving the way for groundbreaking developments in the field of language modeling. Read the research paper, and download the model and code.

AI at Meta

195,607 просмотров • 1 год назад

🚀 Self-speculation brings 6.75x real speedup for LLM generation with SGLang inference! Same model drafts future tokens in Diffusion mode → then verifies them in AR (causal) mode. One model and one KV cache. Just different attention masks. Thanks to perfect alignment, we get 2× longer acceptance lengths than MTP techniques (Eagle-3, MTP, dFlash). We run 2 forward passes… but the 2× higher acceptance means we break even - and with zero overhead from extra drafter, KV cache, or LM head that comes with MTP - those are not free. Last week we released Nemotron-Labs-Diffusion + Tri-mode LLMs! We did continued pre-training on Ministral-3 models by switching attention patterns (block causal bidirectional). Result: one model that runs AR mode, Diffusion mode, and Self-Speculation. Diffusion mode already shows high benchmark accuracy - excited to see what happens when someone beats left-to-right acceptance! 🔥 Github: Paper: SGLang inference: Try the models on HF:

🚀 Self-speculation brings 6.75x real speedup for LLM generation with SGLang inference! Same model drafts future tokens in Diffusion mode → then verifies them in AR (causal) mode. One model and one KV cache. Just different attention masks. Thanks to perfect alignment, we get 2× longer acceptance lengths than MTP techniques (Eagle-3, MTP, dFlash). We run 2 forward passes… but the 2× higher acceptance means we break even - and with zero overhead from extra drafter, KV cache, or LM head that comes with MTP - those are not free. Last week we released Nemotron-Labs-Diffusion + Tri-mode LLMs! We did continued pre-training on Ministral-3 models by switching attention patterns (block causal bidirectional). Result: one model that runs AR mode, Diffusion mode, and Self-Speculation. Diffusion mode already shows high benchmark accuracy - excited to see what happens when someone beats left-to-right acceptance! 🔥 Github: Paper: SGLang inference: Try the models on HF:

Pavlo Molchanov

66,604 просмотров • 2 месяцев назад

Day 11/90 of Inference Engineering How does vLLM work and how is it used in production? Before we discuss how vLLM works internally, it helps to understand what vLLM is. At a high level, vLLM is an inference engine that is designed to serve LLMs to thousands of concurrent users efficiently while managing scarce compute and memory. The goal for vLLM is to maximize throughput and minimize latency; optimizing for the best inference economics and experience for end users. With every request from the end user, it eventually ends up in the engine core, gets scheduled alongside other requests from other concurrent users, executes on the GPU, and updates the KV cache with the new key and value vectors, and streams the tokens back to the user. The Scheduler decides what requests should execute next while continuously batching requests together to maximize GPU utilization. Continuous batching is an inference optimization that allows new requests to join a running batch as other requests finish generating tokens. This helps with keeping the GPU utilization high instead of letting it sit idle waiting for an entire batch to complete generating. After the scheduler dispatches the selected batch to the Model Executor, the Model Executor prepares the tensors and metadata required for inference, retrieves each request’s block table from KV Cache Manager, launches the optimized transformer forward pass on the GPU, computes the logits, updates the KV cache with the new key and value vectors, and finally returns the results for sampling and streaming. The KV Cache Manager uses the PagedAttention memory layout to allocate fixed-size cache blocks on demand and maintains a Free Block Queue on the CPU that tracks which blocks in the GPU’s Paged KV Cache are currently free. When a request needs additional KV cache space, the KV Cache manager takes a free block from the queue and assigns it to that request, thus avoiding an expensive search through GPU memory for available cache blocks. All of these components form the core of vLLM’s inference engine. The Scheduler determines what requests are executed, the Model Executor determines how those requests are executed, the KV Cache Manager determines where each request’s KV cache lives using the PagedAttention Memory Layout. This architecture enables vLLM to serve thousands of concurrent requests with high throughput, low latency, and efficient GPU memory utilization. Heres a little animation that visualizes everything! - I've also completed the forward pass for my mnist.c project. I had a nice chat with shrey birmiwal, such a knowledgeable guy. Excited to learn more about vLLM and implement a tiny-vLLM one day.

Day 11/90 of Inference Engineering How does vLLM work and how is it used in production? Before we discuss how vLLM works internally, it helps to understand what vLLM is. At a high level, vLLM is an inference engine that is designed to serve LLMs to thousands of concurrent users efficiently while managing scarce compute and memory. The goal for vLLM is to maximize throughput and minimize latency; optimizing for the best inference economics and experience for end users. With every request from the end user, it eventually ends up in the engine core, gets scheduled alongside other requests from other concurrent users, executes on the GPU, and updates the KV cache with the new key and value vectors, and streams the tokens back to the user. The Scheduler decides what requests should execute next while continuously batching requests together to maximize GPU utilization. Continuous batching is an inference optimization that allows new requests to join a running batch as other requests finish generating tokens. This helps with keeping the GPU utilization high instead of letting it sit idle waiting for an entire batch to complete generating. After the scheduler dispatches the selected batch to the Model Executor, the Model Executor prepares the tensors and metadata required for inference, retrieves each request’s block table from KV Cache Manager, launches the optimized transformer forward pass on the GPU, computes the logits, updates the KV cache with the new key and value vectors, and finally returns the results for sampling and streaming. The KV Cache Manager uses the PagedAttention memory layout to allocate fixed-size cache blocks on demand and maintains a Free Block Queue on the CPU that tracks which blocks in the GPU’s Paged KV Cache are currently free. When a request needs additional KV cache space, the KV Cache manager takes a free block from the queue and assigns it to that request, thus avoiding an expensive search through GPU memory for available cache blocks. All of these components form the core of vLLM’s inference engine. The Scheduler determines what requests are executed, the Model Executor determines how those requests are executed, the KV Cache Manager determines where each request’s KV cache lives using the PagedAttention Memory Layout. This architecture enables vLLM to serve thousands of concurrent requests with high throughput, low latency, and efficient GPU memory utilization. Heres a little animation that visualizes everything! - I've also completed the forward pass for my mnist.c project. I had a nice chat with shrey birmiwal, such a knowledgeable guy. Excited to learn more about vLLM and implement a tiny-vLLM one day.

max fu

70,543 просмотров • 1 месяц назад

The first natively trained 1-bit model: BitNet 2B. Trained on 4 trillion tokens. that can run on CPUs like Apple M2 Native 1.58-bit weights and 8bit activations W158A8 Outperforms LLaMA &close to Qwen 2.5 1.5B in while using only 0.4GB memory versus 2GB and processes tokens 40%

The first natively trained 1-bit model: BitNet 2B. Trained on 4 trillion tokens. that can run on CPUs like Apple M2 Native 1.58-bit weights and 8bit activations W158A8 Outperforms LLaMA &close to Qwen 2.5 1.5B in while using only 0.4GB memory versus 2GB and processes tokens 40%

Md Ismail Šojal 🕷️

43,786 просмотров • 5 месяцев назад

50% more context unlocked for Qwen 3.8 27b Q4_K_XL dflash 2 on a single RTX 4090 (24 GB VRAM) I found a hidden VRAM tax in llama.cpp. By combining my custom 2 bit DFlash 2 drafter with one overlooked server flag, I just unlocked another +80,000 tokens of context. Qwen3.8-27B is now running a massive 250,000 context at 75 tokens/s on a single RTX 4090. Here is the secret: By default, `llama-server` reserves massive chunks of your VRAM to handle multiple concurrent users (batching). If you are running a single user session, you are bleeding memory for features you aren't using. By passing the `--parallel 1` flag, you force the engine to dedicate 100% of your 24GB VRAM buffer to a single user. When we combine the VRAM saved by our Q2_K 2-bit drafter with the VRAM saved by `--parallel 1`, the context ceilings absolutely explode: Note: all benchmarks carried out with a massive 28k prompt. Ubuntu 22. ### THE NEW 24GB PHYSICAL LIMITS (Single RTX 4090): # 1. The "Repo Swallower" (Q4 KV Cache): - Context: 250,000 tokens (Up from 170k!) - Speed: 73.66 t/s decode | 1,608 t/s prefill - Peak VRAM: 23.8 GB # 2. The "High-Precision SWE" (Q8 KV Cache): - Context: 150,000 tokens (Up from 100k!) - Speed: 75.01 t/s decode | 1,667 t/s prefill - Peak VRAM: 23.9 GB # 3. The "Pristine Attention" (Unquantized FP16 KV): - Context: 90,000 tokens - Speed: 80.58 t/s decode | 1,699 t/s prefill - Peak VRAM: 23.92 GB ### HOW TO RUN THE 250K GOD STACK TODAY: (Requires PR #27342 + my Q2_K Hugging Face drafter) llama.cpp flags: ./build/bin/llama-server -m Qwen3.8-27B-UD-Q4_K_XL.gguf -md Qwen3.8-27B-DFlash2-Q2_K.gguf --spec-type draft-dflash --spec-draft-n-max 3 -c 250000 -ngl 99 --parallel 1 --port 8080 -ctv q4_0 -ctk q4_0 We are pushing a quarter million tokens of context with speculative DFlash 2 decoding at 73 tokens/second on a single consumer gaming GPU. I dropped my custom 2 bit Hugging Face GGUF links, visual performance graphs, and the PR #27342 build instructions in the replies below. If you own a single RTX 3090 or 4090, it is officially time to cancel your API subscriptions and let local silicon eat the cloud. how much monthly API spend does an optimized 4090 rig like this actually replace for you?

50% more context unlocked for Qwen 3.8 27b Q4_K_XL dflash 2 on a single RTX 4090 (24 GB VRAM) I found a hidden VRAM tax in llama.cpp. By combining my custom 2 bit DFlash 2 drafter with one overlooked server flag, I just unlocked another +80,000 tokens of context. Qwen3.8-27B is now running a massive 250,000 context at 75 tokens/s on a single RTX 4090. Here is the secret: By default, `llama-server` reserves massive chunks of your VRAM to handle multiple concurrent users (batching). If you are running a single user session, you are bleeding memory for features you aren't using. By passing the `--parallel 1` flag, you force the engine to dedicate 100% of your 24GB VRAM buffer to a single user. When we combine the VRAM saved by our Q2_K 2-bit drafter with the VRAM saved by `--parallel 1`, the context ceilings absolutely explode: Note: all benchmarks carried out with a massive 28k prompt. Ubuntu 22. ### THE NEW 24GB PHYSICAL LIMITS (Single RTX 4090): # 1. The "Repo Swallower" (Q4 KV Cache): - Context: 250,000 tokens (Up from 170k!) - Speed: 73.66 t/s decode | 1,608 t/s prefill - Peak VRAM: 23.8 GB # 2. The "High-Precision SWE" (Q8 KV Cache): - Context: 150,000 tokens (Up from 100k!) - Speed: 75.01 t/s decode | 1,667 t/s prefill - Peak VRAM: 23.9 GB # 3. The "Pristine Attention" (Unquantized FP16 KV): - Context: 90,000 tokens - Speed: 80.58 t/s decode | 1,699 t/s prefill - Peak VRAM: 23.92 GB ### HOW TO RUN THE 250K GOD STACK TODAY: (Requires PR #27342 + my Q2_K Hugging Face drafter) llama.cpp flags: ./build/bin/llama-server -m Qwen3.8-27B-UD-Q4_K_XL.gguf -md Qwen3.8-27B-DFlash2-Q2_K.gguf --spec-type draft-dflash --spec-draft-n-max 3 -c 250000 -ngl 99 --parallel 1 --port 8080 -ctv q4_0 -ctk q4_0 We are pushing a quarter million tokens of context with speculative DFlash 2 decoding at 73 tokens/second on a single consumer gaming GPU. I dropped my custom 2 bit Hugging Face GGUF links, visual performance graphs, and the PR #27342 build instructions in the replies below. If you own a single RTX 3090 or 4090, it is officially time to cancel your API subscriptions and let local silicon eat the cloud. how much monthly API spend does an optimized 4090 rig like this actually replace for you?

Alok

37,791 просмотров • 2 дней назад

We are excited to introduce Mercury, the first commercial-grade diffusion large language model (dLLM)! dLLMs push the frontier of intelligence and speed with parallel, coarse-to-fine text generation.

We are excited to introduce Mercury, the first commercial-grade diffusion large language model (dLLM)! dLLMs push the frontier of intelligence and speed with parallel, coarse-to-fine text generation.

Inception

1,916,613 просмотров • 1 год назад

LLaDA (the first Large Language Diffusion Model) is *just* out 💥 and I've built a demo, try out now 👨‍💻 It's mesmerizing to watch the diffusion process 🌀, and it being a diffusion model gives you superpowers like "the 4th word has to be pineapple" 🦸 Demo and weights 👇

LLaDA (the first Large Language Diffusion Model) is just out 💥 and I've built a demo, try out now 👨‍💻 It's mesmerizing to watch the diffusion process 🌀, and it being a diffusion model gives you superpowers like "the 4th word has to be pineapple" 🦸 Demo and weights 👇

apolinario (poli)

82,599 просмотров • 1 год назад

You don’t have to use one model (or one provider!) for everything. With Workshop, you can combine frontier and local models in the same workflow. For example: Opus can be the main agent, and delegate specific tasks to Gemma 4 via subagents. Better quality where it matters. Better privacy, speed, and cost where it counts. One workflow, best model for each task.

You don’t have to use one model (or one provider!) for everything. With Workshop, you can combine frontier and local models in the same workflow. For example: Opus can be the main agent, and delegate specific tasks to Gemma 4 via subagents. Better quality where it matters. Better privacy, speed, and cost where it counts. One workflow, best model for each task.

Workshop AI

962,487 просмотров • 4 месяцев назад

Our model continues to be aggressive an wet with the Arthur remnants Thursday night / Friday morning. I still think it's a bit much, but the longer it stays like this the more believable it gets. Main timing we're watching is Thursday night / Friday morning.

Our model continues to be aggressive an wet with the Arthur remnants Thursday night / Friday morning. I still think it's a bit much, but the longer it stays like this the more believable it gets. Main timing we're watching is Thursday night / Friday morning.

Tim Buckley

33,959 просмотров • 2 месяцев назад

Day 12/90 of Inference Engineering What is chunked prefill within vLLM? In continuation of yesterday's post on the high level architecture of vLLM, I want to dive deeper into vLLM core engine starting with the mechanics of chunked prefill. In this post, I will closely follow the original blog on the anatomy of vLLM. To start, let's define chunked prefill. It's a runtime inference optimization technique that splits a long input request so that it doesn’t monopolize the whole GPU. Keep in mind this is all within the context of vLLM. And since vLLM is an inference engine that's meant to serve a model to multiple concurrent users, having a GPU that’s fully monopolized on a single user's request means other users' requests would be in queue waiting to be processed. It isn’t too good to have the whole GPU occupied on a single request when the GPU is meant to be shared! So the key idea behind chunked prefill is to break the long request into smaller chunks, so that each chunk along with other users' requests gets processed and written into the KV cache together. Suppose we split up the long request into chunks and each chunk has 8 tokens. Now each memory block can hold 4 tokens. Therefore, 8 tokens can fit into 2 blocks of memory. After the first forward pass, 2 blocks are occupied, and after the second forward pass, 4 blocks of memory are occupied and so forth. Each forward pass handles a small chunk of the long request so that there's room in the same pass to keep serving other users' requests. Here's a small animation that I made today to fully visualize the idea behind chunked prefill when learning this topic~

Day 12/90 of Inference Engineering What is chunked prefill within vLLM? In continuation of yesterday's post on the high level architecture of vLLM, I want to dive deeper into vLLM core engine starting with the mechanics of chunked prefill. In this post, I will closely follow the original blog on the anatomy of vLLM. To start, let's define chunked prefill. It's a runtime inference optimization technique that splits a long input request so that it doesn’t monopolize the whole GPU. Keep in mind this is all within the context of vLLM. And since vLLM is an inference engine that's meant to serve a model to multiple concurrent users, having a GPU that’s fully monopolized on a single user's request means other users' requests would be in queue waiting to be processed. It isn’t too good to have the whole GPU occupied on a single request when the GPU is meant to be shared! So the key idea behind chunked prefill is to break the long request into smaller chunks, so that each chunk along with other users' requests gets processed and written into the KV cache together. Suppose we split up the long request into chunks and each chunk has 8 tokens. Now each memory block can hold 4 tokens. Therefore, 8 tokens can fit into 2 blocks of memory. After the first forward pass, 2 blocks are occupied, and after the second forward pass, 4 blocks of memory are occupied and so forth. Each forward pass handles a small chunk of the long request so that there's room in the same pass to keep serving other users' requests. Here's a small animation that I made today to fully visualize the idea behind chunked prefill when learning this topic~

max fu

29,417 просмотров • 1 месяц назад

A peanut-sized Chinese model just dethroned Gemini at reading documents. GLM-OCR is a 0.9B parameter vision-language model. It scores 94.62 on OmniDocBench V1.5, ranking #1 overall. For context, it outperforms models 100x its size. 100% open-source. It works in two stages. 1. A layout engine detects every region in a document. 2. Each region gets read in parallel. The model predicts multiple tokens per step instead of one. That's what makes it so fast at small size. It handles things most OCR tools struggle with: > Complex tables and nested layouts > Handwritten text and stamps > Math formulas and code blocks > Mixed image-and-text documents You can run it locally through Ollama. It fits on edge devices with limited compute. Every expensive OCR API just got a free competitor.

A peanut-sized Chinese model just dethroned Gemini at reading documents. GLM-OCR is a 0.9B parameter vision-language model. It scores 94.62 on OmniDocBench V1.5, ranking #1 overall. For context, it outperforms models 100x its size. 100% open-source. It works in two stages. 1. A layout engine detects every region in a document. 2. Each region gets read in parallel. The model predicts multiple tokens per step instead of one. That's what makes it so fast at small size. It handles things most OCR tools struggle with: > Complex tables and nested layouts > Handwritten text and stamps > Math formulas and code blocks > Mixed image-and-text documents You can run it locally through Ollama. It fits on edge devices with limited compute. Every expensive OCR API just got a free competitor.

AlphaSignal

92,071 просмотров • 4 месяцев назад

A peanut-sized Chinese model just dethroned Gemini at reading documents. GLM-OCR is a 0.9B parameter vision-language model. It scores 94.62 on OmniDocBench V1.5, ranking #1 overall. For context, it outperforms models 100x its size. 100% open-source. It works in two stages. 1. A layout engine detects every region in a document. 2. Each region gets read in parallel. The model predicts multiple tokens per step instead of one. That's what makes it so fast at small size. It handles things most OCR tools struggle with: > Complex tables and nested layouts > Handwritten text and stamps > Math formulas and code blocks > Mixed image-and-text documents You can run it locally through Ollama. It fits on edge devices with limited compute. Every expensive OCR API just got a free competitor.

A peanut-sized Chinese model just dethroned Gemini at reading documents. GLM-OCR is a 0.9B parameter vision-language model. It scores 94.62 on OmniDocBench V1.5, ranking #1 overall. For context, it outperforms models 100x its size. 100% open-source. It works in two stages. 1. A layout engine detects every region in a document. 2. Each region gets read in parallel. The model predicts multiple tokens per step instead of one. That's what makes it so fast at small size. It handles things most OCR tools struggle with: > Complex tables and nested layouts > Handwritten text and stamps > Math formulas and code blocks > Mixed image-and-text documents You can run it locally through Ollama. It fits on edge devices with limited compute. Every expensive OCR API just got a free competitor.

Jafar Najafov

13,630 просмотров • 4 месяцев назад

OPEN ZOE VTUBER MODEL We prepared a Zoe Vtuber Model for everyone to use It. So if you want stream as Zoe… be our guest! Link on the next tweet of the thread. We ask you not to do a bad use of it, trusting people will honor the inclusive&thoughtful nature of the franchise

OPEN ZOE VTUBER MODEL We prepared a Zoe Vtuber Model for everyone to use It. So if you want stream as Zoe… be our guest! Link on the next tweet of the thread. We ask you not to do a bad use of it, trusting people will honor the inclusive&thoughtful nature of the franchise

Monster Con is OUT NOW!

53,553 просмотров • 1 год назад

I wanted to test a walten files model, by The Toymaker (Starz) very cool model and it was fun to draw background things even tho i just wanted to test the model the walten files stlye is so cool dude, i didn't know how to portay it in here lol #TheWaltenFiles #blender

I wanted to test a walten files model, by The Toymaker (Starz) very cool model and it was fun to draw background things even tho i just wanted to test the model the walten files stlye is so cool dude, i didn't know how to portay it in here lol #TheWaltenFiles #blender

𝗘𝗟𝗙𝗭𝗚𝗔𝗠𝗘𝗥 On break

166,726 просмотров • 9 дней назад

1/ Gemini 2.5 is here, and it’s our most intelligent AI model ever. Our first 2.5 model, Gemini 2.5 Pro Experimental is a state-of-the-art thinking model, leading in a wide range of benchmarks – with impressive improvements in enhanced reasoning and coding and now #1 on Arena by a significant margin. With a model this intelligent, we wanted to get it to people as quickly as possible. Find it on Google AI Studio and in the Google Gemini for Gemini Advanced users now – and in Vertex in the coming weeks. This is the start of a new era of thinking models – and we can’t wait to see where things go from here.

1/ Gemini 2.5 is here, and it’s our most intelligent AI model ever. Our first 2.5 model, Gemini 2.5 Pro Experimental is a state-of-the-art thinking model, leading in a wide range of benchmarks – with impressive improvements in enhanced reasoning and coding and now #1 on Arena by a significant margin. With a model this intelligent, we wanted to get it to people as quickly as possible. Find it on Google AI Studio and in the Google Gemini for Gemini Advanced users now – and in Vertex in the coming weeks. This is the start of a new era of thinking models – and we can’t wait to see where things go from here.

Sundar Pichai

864,579 просмотров • 1 год назад

the fact that i can take an image of a room and turn it into a 3d model in one shot is actually insane this took like 30 seconds from image to 3d model

the fact that i can take an image of a room and turn it into a 3d model in one shot is actually insane this took like 30 seconds from image to 3d model

Jan

131,884 просмотров • 9 месяцев назад

All the big language models under one roof for the very first time 🤯 Compare the output of OpenAI ChatGPT, Anthropic's Claude and Cohere's language model in a single playground!! Check out this amazing tool to get the best of large language models 👇

All the big language models under one roof for the very first time 🤯 Compare the output of OpenAI ChatGPT, Anthropic's Claude and Cohere's language model in a single playground!! Check out this amazing tool to get the best of large language models 👇

Shubham Saboo

299,259 просмотров • 3 лет назад