Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

introducing simple-llm: a ~950 line, powerful & extensible inference engine that performs on par with vllm. enjoy :) performance (gpt-oss-120b, on an h100): - batch=1: 135 tok/s (vllm: 138) - batch=64: 4,041 tok/s (vllm: 3,846)

naklecha

16,578 subscribers

68,612 görüntüleme • 8 ay önce •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Demo for the sharded inference engine on: gpt-oss-120b on 4090's, max input context 95k, max output context infinite (tok/s degrades with context) queue of max 12 prompts concurrently, use it wisely

Demo for the sharded inference engine on: gpt-oss-120b on 4090's, max input context 95k, max output context infinite (tok/s degrades with context) queue of max 12 prompts concurrently, use it wisely

leyten

90,997 görüntüleme • 2 ay önce

Gemma 4 Diffusion landed in vLLM last week. Day 0. First diffusion LLM natively supported in vLLM. Instead of one token at a time, it predicts 256 tokens at once and iteratively denoises them in parallel. Result: 1,000+ tokens per second at batch size 1 on a single H100. Built on Model Runner V2. Google Gemma

Gemma 4 Diffusion landed in vLLM last week. Day 0. First diffusion LLM natively supported in vLLM. Instead of one token at a time, it predicts 256 tokens at once and iteratively denoises them in parallel. Result: 1,000+ tokens per second at batch size 1 on a single H100. Built on Model Runner V2. Google Gemma

Red Hat AI

17,670 görüntüleme • 3 ay önce

16 people chatting with Qwen3.6-35B at once ONE DGX Spark. This is a real capture, not a mockup: every token you see replays at its actual measured arrival time. Peaks ~440 tok/s total. A single user gets 105 tok/s. NVFP4 + MTP-3 on vLLM.

16 people chatting with Qwen3.6-35B at once ONE DGX Spark. This is a real capture, not a mockup: every token you see replays at its actual measured arrival time. Peaks ~440 tok/s total. A single user gets 105 tok/s. NVFP4 + MTP-3 on vLLM.

Wësche

26,094 görüntüleme • 2 ay önce

Dozens of teams have asked my advice on running LLMs. How fast is DeepSeek V3 with vLLM on 8 GPUs? What's the max throughput of Qwen 2.5 Coder with SGLang on one H100? Running & sharing benchmarks ad hoc was too slow So we built a tiny app, the LLM Engine Advisor

Dozens of teams have asked my advice on running LLMs. How fast is DeepSeek V3 with vLLM on 8 GPUs? What's the max throughput of Qwen 2.5 Coder with SGLang on one H100? Running & sharing benchmarks ad hoc was too slow So we built a tiny app, the LLM Engine Advisor

Charles 🎉 Frye

86,412 görüntüleme • 1 yıl önce

Ran Google’s cookbook with 10 agents on my tiny GB10 GPU. 436 tok/s / 43.6 per agent Qwen3.6-35B + Dflash + DDTree on vLLM GB10 @ 74W The future isn't 10,000 GPUs in a nuclear-powered data center. It’s 10 agents on your desk solving your problems while you make your coffee.

Ran Google’s cookbook with 10 agents on my tiny GB10 GPU. 436 tok/s / 43.6 per agent Qwen3.6-35B + Dflash + DDTree on vLLM GB10 @ 74W The future isn't 10,000 GPUs in a nuclear-powered data center. It’s 10 agents on your desk solving your problems while you make your coffee.

Mitko Vasilev

146,363 görüntüleme • 4 ay önce

DUAL INTEL ARC PRO B70 GPUS AT $1,900 RUN QWEN 3 CODER 30B AT 268 TOKENS PER SECOND VIA VLLM ON 8 CONCURRENT REQUESTS, A NEW INTEL TIER UNDER YOUR MAP THAT BEATS RTX 3090 STACKS ON THROUGHPUT PER DOLLAR 02:22 the operator reads off his benchmark sheet, "vllm was at 268 tokens per second which is just wild to me, getting 250+ tokens per second out of these overall kind of slow Intel cards" dual Intel Arc Pro B70 ships at $950 per card, 32GB GDDR6 each, 64GB combined VRAM for $1,900 total, the same budget gets you two used RTX 3090s at 48GB combined and worse multi-user throughput Qwen 3 coder 30B on dual B70 with vLLM hits 268 t/s at 8 concurrent users, llama.cpp tops out at 84 t/s on the same hardware, the gap is not 2x or 3x, vLLM with Intel's tweaks pulls 3.2x on parallel inference prefill on a 16,000 word context took 26 seconds on llama.cpp and 3 seconds on vLLM, agents spawning subagents or households sharing one box turn that 8x speedup into 250 t/s of usable throughput your map skips Intel because the Arc cards looked weak in 2025, the B70 plus the vLLM stack flipped that, $1,900 in dual Intel silicon now serves a 4 person team at the speed openai serves enterprise tiers for $400 a month per seat bookmark this and read the article below

DUAL INTEL ARC PRO B70 GPUS AT $1,900 RUN QWEN 3 CODER 30B AT 268 TOKENS PER SECOND VIA VLLM ON 8 CONCURRENT REQUESTS, A NEW INTEL TIER UNDER YOUR MAP THAT BEATS RTX 3090 STACKS ON THROUGHPUT PER DOLLAR 02:22 the operator reads off his benchmark sheet, "vllm was at 268 tokens per second which is just wild to me, getting 250+ tokens per second out of these overall kind of slow Intel cards" dual Intel Arc Pro B70 ships at $950 per card, 32GB GDDR6 each, 64GB combined VRAM for $1,900 total, the same budget gets you two used RTX 3090s at 48GB combined and worse multi-user throughput Qwen 3 coder 30B on dual B70 with vLLM hits 268 t/s at 8 concurrent users, llama.cpp tops out at 84 t/s on the same hardware, the gap is not 2x or 3x, vLLM with Intel's tweaks pulls 3.2x on parallel inference prefill on a 16,000 word context took 26 seconds on llama.cpp and 3 seconds on vLLM, agents spawning subagents or households sharing one box turn that 8x speedup into 250 t/s of usable throughput your map skips Intel because the Arc cards looked weak in 2025, the B70 plus the vLLM stack flipped that, $1,900 in dual Intel silicon now serves a 4 person team at the speed openai serves enterprise tiers for $400 a month per seat bookmark this and read the article below

starmex

28,923 görüntüleme • 2 ay önce

Qwen3.6 35B-A3B dropped yesterday, so I ran it on 4 GPUs to see how it performs: 🟣 RTX 3090 — 49.78 tok/s, TTFT 852ms 🟡 RTX 4090 — 118.93 tok/s, TTFT 686ms 🟢 RTX 5090 — 160.37 tok/s, TTFT 409ms 🔵 DGX Spark — 59.98 tok/s, TTFT 228ms I went with ollama as the backend because honestly, it's the easiest way for most people to get started. One command, model pulled, done. I used Q4_K_M (24GB) across all four cards. The reason is the 3090 and 4090 don't support NVFP4 (only the 5090 and DGX Spark could use it). Keeping the same quant everywhere felt like the fairest way to compare. And yes, you can absolutely squeeze more performance out of every card with vLLM, SGLang, or TensorRT-LLM. But that's not what this test is about. This is just the out-of-the-box experience for folks who own a GPU and want to try the new model tonight.

Qwen3.6 35B-A3B dropped yesterday, so I ran it on 4 GPUs to see how it performs: 🟣 RTX 3090 — 49.78 tok/s, TTFT 852ms 🟡 RTX 4090 — 118.93 tok/s, TTFT 686ms 🟢 RTX 5090 — 160.37 tok/s, TTFT 409ms 🔵 DGX Spark — 59.98 tok/s, TTFT 228ms I went with ollama as the backend because honestly, it's the easiest way for most people to get started. One command, model pulled, done. I used Q4_K_M (24GB) across all four cards. The reason is the 3090 and 4090 don't support NVFP4 (only the 5090 and DGX Spark could use it). Keeping the same quant everywhere felt like the fairest way to compare. And yes, you can absolutely squeeze more performance out of every card with vLLM, SGLang, or TensorRT-LLM. But that's not what this test is about. This is just the out-of-the-box experience for folks who own a GPU and want to try the new model tonight.

stevibe

403,049 görüntüleme • 4 ay önce

🤯 A localmaxxer hit ~381 tok/s on a SINGLE RTX 3090 with Qwen3.8-27B. This developer has turned a 24GB RTX 3090 into a monster Qwen inference box - w/some creativity. Four days ago 👉 ⚡ ~82 tok/s single-user Then 👉 ⚡ ~114 tok/s with optimized MTP ⚡ ~138 tok/s with DFlash2 + lookup drafting Now 👉 🔥 ~381 tok/s on ONE request How? The recipe combines ... 🧠 Qwen3.8-27B 🎮 1× RTX 3090 24GB @ 250W ⚙️ heavily optimized vLLM ⚡ DFlash2 speculative decoding 🔎 lookup-augmented drafting 📚 prefix caching 🧮 16-token verification blocks 💾 quantized KV / heads / activations DFlash2 normally proposes 7 tokens. The developer realized the verification block doesn't have to stop there. If Qwen is answering from a document already sitting in the prompt, the system can fill the remaining draft positions using tokens found directly in that context. 🎯 So the target model can verify 16 tokens at once. On a ~25K-token document reproduction task: Previous DFlash2 👉 ~260 tok/s Longer verification + context lookup 👉 🔥 ~382 tok/s Acceptance: 🤯 15 of 16 tokens per verification step That is where the crazy number comes from. ⚠️ On ordinary real-world chat prompts, the same setup is around ~133 tok/s Still extremely fast for a dense 27B model on an RTX 3090. The 381 tok/s mode shines when the answer largely comes from material already in context so these are best use cases 📚 RAG / document Q&A 💻 Coding assistants applying edits 📝 Quoting or rewriting documents 🔎 Extracting information from long prompts And another optimization 👉 With prefix caching, a second question against the same 25K-token document reportedly goes from: 🐌 22.4 sec TTFT → ⚡ 0.56 sec TTFT Because the model doesn't need to process the whole document again. 🎯 It's specifically a mode for RAG front ends and coding agents. Follow iamMess on Reddit or syv-ai on GitHub 🔗 Reddit: r/LocalLLaMA/comments/1vtup5s/ 🔗 GitHub: /syv-ai/qwen38-27b-rtx3090

🤯 A localmaxxer hit ~381 tok/s on a SINGLE RTX 3090 with Qwen3.8-27B. This developer has turned a 24GB RTX 3090 into a monster Qwen inference box - w/some creativity. Four days ago 👉 ⚡ ~82 tok/s single-user Then 👉 ⚡ ~114 tok/s with optimized MTP ⚡ ~138 tok/s with DFlash2 + lookup drafting Now 👉 🔥 ~381 tok/s on ONE request How? The recipe combines ... 🧠 Qwen3.8-27B 🎮 1× RTX 3090 24GB @ 250W ⚙️ heavily optimized vLLM ⚡ DFlash2 speculative decoding 🔎 lookup-augmented drafting 📚 prefix caching 🧮 16-token verification blocks 💾 quantized KV / heads / activations DFlash2 normally proposes 7 tokens. The developer realized the verification block doesn't have to stop there. If Qwen is answering from a document already sitting in the prompt, the system can fill the remaining draft positions using tokens found directly in that context. 🎯 So the target model can verify 16 tokens at once. On a ~25K-token document reproduction task: Previous DFlash2 👉 ~260 tok/s Longer verification + context lookup 👉 🔥 ~382 tok/s Acceptance: 🤯 15 of 16 tokens per verification step That is where the crazy number comes from. ⚠️ On ordinary real-world chat prompts, the same setup is around ~133 tok/s Still extremely fast for a dense 27B model on an RTX 3090. The 381 tok/s mode shines when the answer largely comes from material already in context so these are best use cases 📚 RAG / document Q&A 💻 Coding assistants applying edits 📝 Quoting or rewriting documents 🔎 Extracting information from long prompts And another optimization 👉 With prefix caching, a second question against the same 25K-token document reportedly goes from: 🐌 22.4 sec TTFT → ⚡ 0.56 sec TTFT Because the model doesn't need to process the whole document again. 🎯 It's specifically a mode for RAG front ends and coding agents. Follow iamMess on Reddit or syv-ai on GitHub 🔗 Reddit: r/LocalLLaMA/comments/1vtup5s/ 🔗 GitHub: /syv-ai/qwen38-27b-rtx3090

David Hendrickson

113,342 görüntüleme • 23 gün önce

Qwen3.8-Flash-Next just got a massive speed boost on my old hardware. This guy started with roughly 11 tok/s. After tuning the inference stack, I’m now getting around 24–28 tok/s. That’s roughly a 2× speedup without changing the GPU. The setup: → RTX 3090 24GB → 64GB DDR4 system RAM → Qwen3.8-Flash-Next → Cafe-llama.cpp The interesting part is that Cafe-llama.cpp isn’t just another llama.cpp fork. The optimizations are making a real difference for this particular architecture, especially when you’re dealing with a model that relies heavily on system memory and offloading. Going from 11 tok/s to nearly 30 tok/s completely changes how practical the model feels for local use. And now I’m curious what happens with DDR5. If memory bandwidth becomes another bottleneck, a faster RAM subsystem could squeeze even more performance out of the same 3090. This is why local AI optimization is getting so interesting. Sometimes you don’t need a new GPU. You just need a better inference stack. 11 tok/s → 24–28 tok/s.

Qwen3.8-Flash-Next just got a massive speed boost on my old hardware. This guy started with roughly 11 tok/s. After tuning the inference stack, I’m now getting around 24–28 tok/s. That’s roughly a 2× speedup without changing the GPU. The setup: → RTX 3090 24GB → 64GB DDR4 system RAM → Qwen3.8-Flash-Next → Cafe-llama.cpp The interesting part is that Cafe-llama.cpp isn’t just another llama.cpp fork. The optimizations are making a real difference for this particular architecture, especially when you’re dealing with a model that relies heavily on system memory and offloading. Going from 11 tok/s to nearly 30 tok/s completely changes how practical the model feels for local use. And now I’m curious what happens with DDR5. If memory bandwidth becomes another bottleneck, a faster RAM subsystem could squeeze even more performance out of the same 3090. This is why local AI optimization is getting so interesting. Sometimes you don’t need a new GPU. You just need a better inference stack. 11 tok/s → 24–28 tok/s.

FHILY👑

23,243 görüntüleme • 13 gün önce

got gemma 4 31B with MTP running on my DGX Spark. Hermes Agent did most of the legwork. baseline vs MTP on GB10: • c=1: 3.65 → 6.37 tok/s (1.74x) • c=4: 14.34 → 23.59 tok/s (1.65x) • c=8: 14.37 → 24.18 tok/s (1.68x) google says "up to 2x" — we're not quite there but it's real, not vapor. stack: DGX Spark / GB10 + gemma-4-31b-it + gemma-4-31b-it-assistant (MTP drafter) + vLLM built from PR 41745 MTP is basically a lightweight draft model that predicts multiple tokens while the big model verifies them all at once. smaller model does the busywork, bigger model just says yes/no. simple idea, weird to implement. next: tune the draft block size and see if we can push past 2x. also want to try it with Hermes Agent feeding prompts end to end. p.s: this was all done from telegram. Google DeepMind NVIDIA AI Developer

got gemma 4 31B with MTP running on my DGX Spark. Hermes Agent did most of the legwork. baseline vs MTP on GB10: • c=1: 3.65 → 6.37 tok/s (1.74x) • c=4: 14.34 → 23.59 tok/s (1.65x) • c=8: 14.37 → 24.18 tok/s (1.68x) google says "up to 2x" — we're not quite there but it's real, not vapor. stack: DGX Spark / GB10 + gemma-4-31b-it + gemma-4-31b-it-assistant (MTP drafter) + vLLM built from PR 41745 MTP is basically a lightweight draft model that predicts multiple tokens while the big model verifies them all at once. smaller model does the busywork, bigger model just says yes/no. simple idea, weird to implement. next: tune the draft block size and see if we can push past 2x. also want to try it with Hermes Agent feeding prompts end to end. p.s: this was all done from telegram. Google DeepMind NVIDIA AI Developer

Joey

23,088 görüntüleme • 4 ay önce

i spent 3 hours finding the sweet spot for hermes 4.3 36B on a single RTX 3090. saving you the trouble, anon. the model is 21.8GB at Q4_K_M. that leaves 2.2GB free on 24GB VRAM. not much room for KV cache. here's what actually happened: 4K: 35.3 tok/s 8K: 35.2 tok/s 16K: 34.8 tok/s 32K: 34.6 tok/s 64K: 6.4 tok/s 128K: 1.9 tok/s flat from 4K to 32K. then it falls off a cliff at 64K. the trick is quantized KV cache. without it you OOM at 16K. with quantized KV cache you get 32K at full speed. all 64 layers on GPU. at 64K something weird happens. ngl 99 (all layers on GPU) = 3.96 tok/s. the KV cache silently spills to CPU. drop to ngl 55 and speed jumps to 6.37. drop to ngl 48 and it gets worse again (3.46). there's an offload sweet spot where you free just enough VRAM for the cache without losing too much compute to PCIe transfers. 128K works at ngl 32 but you're at 1.95 tok/s. half the model on CPU. usable for batch work, not for interactive. the sweet spot command: llama-server -m hermes-4.3-36b-Q4_K_M.gguf -ngl 99 -c 32768 --cache-type-k q4_0 --cache-type-v q4_0 32K context. 34.6 tok/s. all on GPU. this is where dense 36B lives on 24GB. for comparison, qwen 3.5 (35B MoE, 3B active) holds 112 tok/s from 4K all the way to 262K on the same GPU. no speed drop. same total params, completely different architecture. hybrid linear attention means flat context scaling. dense pays for every token in the KV cache. code and quality comparison coming next. fast vs slow generation side by side in the videos below.

i spent 3 hours finding the sweet spot for hermes 4.3 36B on a single RTX 3090. saving you the trouble, anon. the model is 21.8GB at Q4_K_M. that leaves 2.2GB free on 24GB VRAM. not much room for KV cache. here's what actually happened: 4K: 35.3 tok/s 8K: 35.2 tok/s 16K: 34.8 tok/s 32K: 34.6 tok/s 64K: 6.4 tok/s 128K: 1.9 tok/s flat from 4K to 32K. then it falls off a cliff at 64K. the trick is quantized KV cache. without it you OOM at 16K. with quantized KV cache you get 32K at full speed. all 64 layers on GPU. at 64K something weird happens. ngl 99 (all layers on GPU) = 3.96 tok/s. the KV cache silently spills to CPU. drop to ngl 55 and speed jumps to 6.37. drop to ngl 48 and it gets worse again (3.46). there's an offload sweet spot where you free just enough VRAM for the cache without losing too much compute to PCIe transfers. 128K works at ngl 32 but you're at 1.95 tok/s. half the model on CPU. usable for batch work, not for interactive. the sweet spot command: llama-server -m hermes-4.3-36b-Q4_K_M.gguf -ngl 99 -c 32768 --cache-type-k q4_0 --cache-type-v q4_0 32K context. 34.6 tok/s. all on GPU. this is where dense 36B lives on 24GB. for comparison, qwen 3.5 (35B MoE, 3B active) holds 112 tok/s from 4K all the way to 262K on the same GPU. no speed drop. same total params, completely different architecture. hybrid linear attention means flat context scaling. dense pays for every token in the KV cache. code and quality comparison coming next. fast vs slow generation side by side in the videos below.

Sudo su

52,268 görüntüleme • 6 ay önce

I added KV caching and INT8 KV quantization to our transformer inference, improving throughput by 35x. All of this was done from scratch in Rust + CUDA, on top of a homemade ML framework. On a 4-token prompt with 252 generated tokens: - Original: 0.76 tok/s - KV cache fp32: 27.21 tok/s - KV cache int8 (quantized): 27.29 tok/s Try it out yourself here: In practice: - KV caching gave us about a 35x end-to-end speedup - INT8 KV cache kept roughly the same speed as fp32 but cut KV cache memory by 3.78x FP32 cache used 4.5 MB in this run while the INT8 cache used only 1.19 MB This simple change to inference created a huge impact on performance. To learn more about the KV cache and other optimizations like this, check out the blog at

I added KV caching and INT8 KV quantization to our transformer inference, improving throughput by 35x. All of this was done from scratch in Rust + CUDA, on top of a homemade ML framework. On a 4-token prompt with 252 generated tokens: - Original: 0.76 tok/s - KV cache fp32: 27.21 tok/s - KV cache int8 (quantized): 27.29 tok/s Try it out yourself here: In practice: - KV caching gave us about a 35x end-to-end speedup - INT8 KV cache kept roughly the same speed as fp32 but cut KV cache memory by 3.78x FP32 cache used 4.5 MB in this run while the INT8 cache used only 1.19 MB This simple change to inference created a huge impact on performance. To learn more about the KV cache and other optimizations like this, check out the blog at

Reese Chong

53,026 görüntüleme • 4 ay önce

PowerInfer - a high-speed inference engine for deploying LLMs locally. Just came across this super interesting project on speeding up inference. It's not MoE but it's a simple approach that exploits the high locality in LLM inference to design a GPU-CPU hybrid inference engine. Hot-activated neurons are preloaded onto the GPU for fast access, while cold-activated neurons (the majority) are computed on the CPU. This approach significantly reduces GPU memory demands and CPU-GPU data transfer. It achieves an average token generation rate of 13.20 tokens/s, with a peak of 29.08 tokens/s, across various LLMs on a single NVIDIA RTX 4090 GPU. It's on only 18% lower than that achieved by a top-tier server-grade A100 GPU. It also significantly outperforms llama.cpp by up to 11.69x while retaining model accuracy. There is a lot more innovation around inference that's coming fast. Really encouraged by the study on sparse computation to enhance the computational efficiency of LLMs. It's now possible to use PowerInfer with Llama 2 and Faclon 40B. Mistral-7B support is coming soon!

PowerInfer - a high-speed inference engine for deploying LLMs locally. Just came across this super interesting project on speeding up inference. It's not MoE but it's a simple approach that exploits the high locality in LLM inference to design a GPU-CPU hybrid inference engine. Hot-activated neurons are preloaded onto the GPU for fast access, while cold-activated neurons (the majority) are computed on the CPU. This approach significantly reduces GPU memory demands and CPU-GPU data transfer. It achieves an average token generation rate of 13.20 tokens/s, with a peak of 29.08 tokens/s, across various LLMs on a single NVIDIA RTX 4090 GPU. It's on only 18% lower than that achieved by a top-tier server-grade A100 GPU. It also significantly outperforms llama.cpp by up to 11.69x while retaining model accuracy. There is a lot more innovation around inference that's coming fast. Really encouraged by the study on sparse computation to enhance the computational efficiency of LLMs. It's now possible to use PowerInfer with Llama 2 and Faclon 40B. Mistral-7B support is coming soon!

elvis

261,622 görüntüleme • 2 yıl önce

Announcing Cloudy, a platform that seamlessly handles your training infrastructure. You can rent a single H100 or a cluster of 1000 H100s, manage petabyte-scale storage volumes & seamlessly go from running experiments to managing large scale training runs on a single interface.

Announcing Cloudy, a platform that seamlessly handles your training infrastructure. You can rent a single H100 or a cluster of 1000 H100s, manage petabyte-scale storage volumes & seamlessly go from running experiments to managing large scale training runs on a single interface.

naklecha

33,343 görüntüleme • 11 ay önce

Timelapse #156 (36 hrs) - Worked with the tiny corp on getting GLM 5.2 running on 8xMI300X (sglang won here) - Launched KernelBench-Mega and updated Kernelbench-Hard with h100 and b200 sweeps - Took care of boring business stuff - Did some training sweeps for specialized technical vocab audio model - Some bugs with putting kernelbench-mega and hard on cloud instances so had to do some reruns. Learned a lot though - Setting up my own local rl infra and profiling concurrency 128 rollouts with vllm. Became clear to me that I need to serve in nvfp4, use MoE only for throughput, reap so training doesn’t OOM, dig into the vllm kernel graph itself to not underutilize my hardware from poor flashinfer/cutlass selections for my rtx pro 6000 sm120 architecture - Might do online distillation from glm 5.2 but for now taking it one step at a time - Slept for a bit then woke up and showered - Fixed an issue with SGLang tensor parallel deadlock on GLM 5.2 architecture with MTP enabled - GLM 5.2 inference is 2-3x faster than coding plans and running on amd boxes - Spent time with family - Hung out with some friends - Recorded some yoctogpt lectures with the revamped notebook (high taste btw) - Setting up dflash training for GLM 5.2

Timelapse #156 (36 hrs) - Worked with the tiny corp on getting GLM 5.2 running on 8xMI300X (sglang won here) - Launched KernelBench-Mega and updated Kernelbench-Hard with h100 and b200 sweeps - Took care of boring business stuff - Did some training sweeps for specialized technical vocab audio model - Some bugs with putting kernelbench-mega and hard on cloud instances so had to do some reruns. Learned a lot though - Setting up my own local rl infra and profiling concurrency 128 rollouts with vllm. Became clear to me that I need to serve in nvfp4, use MoE only for throughput, reap so training doesn’t OOM, dig into the vllm kernel graph itself to not underutilize my hardware from poor flashinfer/cutlass selections for my rtx pro 6000 sm120 architecture - Might do online distillation from glm 5.2 but for now taking it one step at a time - Slept for a bit then woke up and showered - Fixed an issue with SGLang tensor parallel deadlock on GLM 5.2 architecture with MTP enabled - GLM 5.2 inference is 2-3x faster than coding plans and running on amd boxes - Spent time with family - Hung out with some friends - Recorded some yoctogpt lectures with the revamped notebook (high taste btw) - Setting up dflash training for GLM 5.2

Elliot Arledge

45,447 görüntüleme • 2 ay önce

Continuous batching in LLMs, clearly explained: (a popular LLM interview question; bookmark this) In traditional ML inference, a batch is a matrix. Every input is padded to the same length, one forward pass runs, and every row finishes at the same moment. LLM decoding does not work that way. One forward pass produces one token per sequence, so a request needs as many passes as it has output tokens, and nobody knows that count until the model emits a stop token. Under static batching, membership is fixed when the batch starts. A request that finishes in 30 tokens holds its slot until the slowest request in the same batch finishes at 400. The GPU keeps paying the full weight read for a batch that is mostly empty. Loading model weights out of HBM costs the same whether four slots are producing tokens or one. Continuous batching moves the decision boundary. Instead of scheduling once per batch, the scheduler runs a single forward pass, gets control back, and decides again. A finished request leaves at the next iteration boundary, and a queued request takes its slot right there. No slot stays reserved for work that is already done. Anyscale benchmarked both OPT-13B on a single A100. With uniform generation lengths, the two policies came out about level (as expected), and as output length variance rose, static batching fell to around 81 tokens per second while vLLM reached 23x the throughput of naive Hugging Face serving. Variance drives the entire gap. Production traffic mixes 30-token replies with 400-token ones, which is exactly the condition static batching handles worst. None of this alters the model. vLLM, SGLang, TGI, and TensorRT-LLM all run it by default, and NVIDIA ships the same mechanism under the name in-flight batching. The animation below runs both policies on the same 16 requests and the same 4 slots, stepping in lockstep. The only difference is when a new request is allowed in. To dive deeper into continuous batching specifically, I wrote a full breakdown of the scheduler underneath it. It covers what happens between two forward passes, how tokens get handed out against a fixed budget, why the scheduler needs no separate path for prefill and decode, and what preemption costs you when the KV cache fills up mid-generation. Read it below.

Continuous batching in LLMs, clearly explained: (a popular LLM interview question; bookmark this) In traditional ML inference, a batch is a matrix. Every input is padded to the same length, one forward pass runs, and every row finishes at the same moment. LLM decoding does not work that way. One forward pass produces one token per sequence, so a request needs as many passes as it has output tokens, and nobody knows that count until the model emits a stop token. Under static batching, membership is fixed when the batch starts. A request that finishes in 30 tokens holds its slot until the slowest request in the same batch finishes at 400. The GPU keeps paying the full weight read for a batch that is mostly empty. Loading model weights out of HBM costs the same whether four slots are producing tokens or one. Continuous batching moves the decision boundary. Instead of scheduling once per batch, the scheduler runs a single forward pass, gets control back, and decides again. A finished request leaves at the next iteration boundary, and a queued request takes its slot right there. No slot stays reserved for work that is already done. Anyscale benchmarked both OPT-13B on a single A100. With uniform generation lengths, the two policies came out about level (as expected), and as output length variance rose, static batching fell to around 81 tokens per second while vLLM reached 23x the throughput of naive Hugging Face serving. Variance drives the entire gap. Production traffic mixes 30-token replies with 400-token ones, which is exactly the condition static batching handles worst. None of this alters the model. vLLM, SGLang, TGI, and TensorRT-LLM all run it by default, and NVIDIA ships the same mechanism under the name in-flight batching. The animation below runs both policies on the same 16 requests and the same 4 slots, stepping in lockstep. The only difference is when a new request is allowed in. To dive deeper into continuous batching specifically, I wrote a full breakdown of the scheduler underneath it. It covers what happens between two forward passes, how tokens get handed out against a fixed budget, why the scheduler needs no separate path for prefill and decode, and what preemption costs you when the KV cache fills up mid-generation. Read it below.

Avi Chawla

16,140 görüntüleme • 1 ay önce

"Mac Minis for example are a very good fit" - Andrej Karpathy Andrej Karpathy shouted out my work on EXO Labs in his keynote at Y Combinator AI SUS! Here's the breakdown: Right now most AI workloads run in the cloud where requests from different users are continuously batched together. These workloads are FLOPS-bound and favors hardware with the best unit economics of $ per FLOP, i.e. enterprise GPUs. The personal computing revolution will shift these workloads to personal devices with lower batch sizes (mostly batch_size=1). batch_size=1 inference is memory-bound, because all of the model parameters need to be loaded into the GPU every time a token is generated. Apple Silicon with its Unified Memory architecture has a lot of memory and memory bandwidth per $ compared to other hardware: - M4 Pro Mac Mini, 24GB @ 273GB/s, $58.33/GB, $5.13/GB/s - H100, 80GB @ 3350GB/s, $625/GB, $14.93/GB/s The unit economics of Apple Silicon are becoming more compelling with every release of the Mac. The future of AI inference looks more like open weights models (OpenAI open weights model soonTM) run at low batch_size on personal devices.

"Mac Minis for example are a very good fit" - Andrej Karpathy Andrej Karpathy shouted out my work on EXO Labs in his keynote at Y Combinator AI SUS! Here's the breakdown: Right now most AI workloads run in the cloud where requests from different users are continuously batched together. These workloads are FLOPS-bound and favors hardware with the best unit economics of $ per FLOP, i.e. enterprise GPUs. The personal computing revolution will shift these workloads to personal devices with lower batch sizes (mostly batch_size=1). batch_size=1 inference is memory-bound, because all of the model parameters need to be loaded into the GPU every time a token is generated. Apple Silicon with its Unified Memory architecture has a lot of memory and memory bandwidth per $ compared to other hardware: - M4 Pro Mac Mini, 24GB @ 273GB/s, $58.33/GB, $5.13/GB/s - H100, 80GB @ 3350GB/s, $625/GB, $14.93/GB/s The unit economics of Apple Silicon are becoming more compelling with every release of the Mac. The future of AI inference looks more like open weights models (OpenAI open weights model soonTM) run at low batch_size on personal devices.

Alex Cheema

103,604 görüntüleme • 1 yıl önce

🎉 Congrats to Fish Audio on launching Fish Audio S2, a frontier TTS model with fine-grained prosody & emotion control via natural-language inline tags. SGLang Day-0 support is now live! 🏆 Best WER on Seed-TTS Eval; 81.88% win rate on EmergentTTS-Eval 🎙️ Voice cloning with 86.4% prefix-cache hit rate via RadixAttention ⚡️ RTF 0.34, 63.3 tok/s on single H200 (single batch) 🌍 Trained on 10M+ hours of audio across ~100 languages, GRPO-aligned 🔧 Dual-AR (Slow + Fast AR) is LLM-isomorphic: continuous batching, paged KV cache & CUDA graphs inherited natively 🗣️ Native multi-speaker: turn-taking, interruptions & cross-speaker emotion in a single pass 👉Cookbook: 👉Blog: 🎬 Curious how to run with SGLang? Check out this voice cloning demo from Chayenne Zhao with Fishaudio-S2-Pro:

🎉 Congrats to Fish Audio on launching Fish Audio S2, a frontier TTS model with fine-grained prosody & emotion control via natural-language inline tags. SGLang Day-0 support is now live! 🏆 Best WER on Seed-TTS Eval; 81.88% win rate on EmergentTTS-Eval 🎙️ Voice cloning with 86.4% prefix-cache hit rate via RadixAttention ⚡️ RTF 0.34, 63.3 tok/s on single H200 (single batch) 🌍 Trained on 10M+ hours of audio across ~100 languages, GRPO-aligned 🔧 Dual-AR (Slow + Fast AR) is LLM-isomorphic: continuous batching, paged KV cache & CUDA graphs inherited natively 🗣️ Native multi-speaker: turn-taking, interruptions & cross-speaker emotion in a single pass 👉Cookbook: 👉Blog: 🎬 Curious how to run with SGLang? Check out this voice cloning demo from Chayenne Zhao with Fishaudio-S2-Pro:

LMSYS Org

39,939 görüntüleme • 6 ay önce

Researchers found a way to make LLMs 8.5x faster! (without compromising accuracy) Speculative decoding is quite an effective way to address the single-token bottleneck in traditional LLM inference. A small "draft" model first generates the next several tokens, then the large model verifies all of them at once in a single forward pass. If a token at any position is wrong, you keep everything before it and restart from there. This never does worse than normal decoding. But current drafters in Speculative decoding still guess one token at a time. That makes the drafting step itself a bottleneck, capping real-world speedups at 2-3x. DFlash is a new technique that swaps the autoregressive drafter with a lightweight block diffusion model that guesses all tokens in one parallel shot. Drafting cost stays flat no matter how many tokens you speculate. On top of that, the drafter is conditioned on hidden features pulled from multiple layers of the target model and injected into every draft layer, so it makes significantly better guesses than a drafter working from scratch. In the side-by-side demo below, vanilla decoding runs at 48.5 tokens/sec. DFlash hits 415 tokens/sec on the same model, with zero quality loss. It's already integrated with vLLM, SGLang, and Transformers, with draft models on HuggingFace for several models like Qwen3, Qwen3.5, Llama 3.1, Kimi-K2.5, gpt-oss, and many more. I have shared the GitHub repo in the replies! KV caching is another must-know technique to boost LLM inference. I recently wrote an article about it. Read it below. 👉 Over to you: What use case are you working on that can benefit from this new technique?

Researchers found a way to make LLMs 8.5x faster! (without compromising accuracy) Speculative decoding is quite an effective way to address the single-token bottleneck in traditional LLM inference. A small "draft" model first generates the next several tokens, then the large model verifies all of them at once in a single forward pass. If a token at any position is wrong, you keep everything before it and restart from there. This never does worse than normal decoding. But current drafters in Speculative decoding still guess one token at a time. That makes the drafting step itself a bottleneck, capping real-world speedups at 2-3x. DFlash is a new technique that swaps the autoregressive drafter with a lightweight block diffusion model that guesses all tokens in one parallel shot. Drafting cost stays flat no matter how many tokens you speculate. On top of that, the drafter is conditioned on hidden features pulled from multiple layers of the target model and injected into every draft layer, so it makes significantly better guesses than a drafter working from scratch. In the side-by-side demo below, vanilla decoding runs at 48.5 tokens/sec. DFlash hits 415 tokens/sec on the same model, with zero quality loss. It's already integrated with vLLM, SGLang, and Transformers, with draft models on HuggingFace for several models like Qwen3, Qwen3.5, Llama 3.1, Kimi-K2.5, gpt-oss, and many more. I have shared the GitHub repo in the replies! KV caching is another must-know technique to boost LLM inference. I recently wrote an article about it. Read it below. 👉 Over to you: What use case are you working on that can benefit from this new technique?

Avi Chawla

157,390 görüntüleme • 4 ay önce

SGLang now supports DSpark, enabling confidence-driven, variable-length verification for speculative decoding 🎉 DSpark addresses a key bottleneck under load: instead of verifying every draft token, it verifies only where the draft model is confident, so the gains hold even as batch size scales. We heavily optimized variable-length verification in SGLang. Across batch sizes 1 to 256, DSpark gives the best throughput/latency tradeoff on DeepSeek-V4-Flash, ahead of both MTP and non-spec. At high concurrency, dynamic scheduling provides up to ~20% higher throughput compared to a fixed budget, while maintaining high verification quality across workloads. With fused kernels and zero-overhead scheduling, DeepSeek-V4-Pro reaches 383.7 tok/s at B=1 on B300. DSpark is now available in SGLang with support for Qwen3 and DeepSeek-V4. Thanks DeepSeek for open-sourcing! Blog with full technical details and commands to run below 👇

SGLang now supports DSpark, enabling confidence-driven, variable-length verification for speculative decoding 🎉 DSpark addresses a key bottleneck under load: instead of verifying every draft token, it verifies only where the draft model is confident, so the gains hold even as batch size scales. We heavily optimized variable-length verification in SGLang. Across batch sizes 1 to 256, DSpark gives the best throughput/latency tradeoff on DeepSeek-V4-Flash, ahead of both MTP and non-spec. At high concurrency, dynamic scheduling provides up to ~20% higher throughput compared to a fixed budget, while maintaining high verification quality across workloads. With fused kernels and zero-overhead scheduling, DeepSeek-V4-Pro reaches 383.7 tok/s at B=1 on B300. DSpark is now available in SGLang with support for Qwen3 and DeepSeek-V4. Thanks DeepSeek for open-sourcing! Blog with full technical details and commands to run below 👇

LMSYS Org

170,400 görüntüleme • 2 ay önce