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.

Did you know Qwen 3.6 35B NVFP4 with 256k context, MTP, kv cache fp16 & image support takes around 68gb of VRAM? My NVIDIA AI DGX Spark can do 110 tok/s 👇

Mia

7,889 subscribers

35,255 views • 1 month 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

One DGX Spark. Qwen3.6-35B 64 users. 700+ tok/s. 32,768 tokens in 54 seconds. 38W Each user has their own prompt and their own KV cache, and vLLM batches every active stream through the GPU each step. Recipe → NVIDIA AI

One DGX Spark. Qwen3.6-35B 64 users. 700+ tok/s. 32,768 tokens in 54 seconds. 38W Each user has their own prompt and their own KV cache, and vLLM batches every active stream through the GPU each step. Recipe → NVIDIA AI

Wësche

164,914 views • 1 month ago

"Why are you benchmarking DGX Spark? It's a training box." Yeah. Low bandwidth, but 128GB of unified memory is just sitting there. Plenty of room to optimize. DGX Spark + Qwen3.6 27B. Four backend/quant combos: 🔴 llama.cpp + UD_Q4_K_XL > 11.0 tok/s (baseline), TTFT 297ms 🟢 llama.cpp + DFlash > 20.4 tok/s (peaks at 97 tok/s), TTFT 320ms 🟡 vLLM FP8 + MTP > 13.1 tok/s, TTFT 540ms 🟣 vLLM NVFP4 + MTP > 24.2 tok/s, TTFT 376ms NVFP4+MTP is the winner for me, rock stable around 24 tok/s, no wild swings. DFlash is the wildcard: massive peaks, but fluctuates a lot. FP8+MTP barely beats baseline, and it's FP8. Love my Spark.

"Why are you benchmarking DGX Spark? It's a training box." Yeah. Low bandwidth, but 128GB of unified memory is just sitting there. Plenty of room to optimize. DGX Spark + Qwen3.6 27B. Four backend/quant combos: 🔴 llama.cpp + UD_Q4_K_XL > 11.0 tok/s (baseline), TTFT 297ms 🟢 llama.cpp + DFlash > 20.4 tok/s (peaks at 97 tok/s), TTFT 320ms 🟡 vLLM FP8 + MTP > 13.1 tok/s, TTFT 540ms 🟣 vLLM NVFP4 + MTP > 24.2 tok/s, TTFT 376ms NVFP4+MTP is the winner for me, rock stable around 24 tok/s, no wild swings. DFlash is the wildcard: massive peaks, but fluctuates a lot. FP8+MTP barely beats baseline, and it's FP8. Love my Spark.

stevibe

40,118 views • 3 months ago

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 views • 1 month ago

Hot off the press 🔥 AEON vLLM Ultimate → v0.24.0. Link below ⤵️ The ultimate container for AI on NVIDIA DGX Spark — vLLM compiled from source for GB10/sm_121a. 🔓 UNLOCKED (impossible until this release): ▸ DFlash + FP8 KV cache — 2× KV capacity, draft acceptance intact, measured on-device ▸ FP8 KV on vision models — the KV-bandwidth lever is back ▸ DFlash on FlashInfer & Triton backends (was FlashAttention-only) 🛠 Carried ahead of upstream (still-open PRs, merged + validated): ▸ Triton NVFP4 KV cache → up to 3× KV capacity ▸ DFlash sliding-window drafters + prefix-cache stability for DFlash ▸ High-concurrency DFlash — clean through c=64 🧰 Fixes stock v0.24.0 doesn't have: ▸ Gemma-4 tied embeddings (ModelOpt) — stock crashes at load, ours doesn't ▸ Tool calling out of the box (stock --no-deps builds 500) ▸ Unified-memory hardened — no silent KV overcommit on the 121 GB pool 📊 One image, the whole fleet: Gemma-4-26B-A4B · Qwen3.6-27B · Qwen3.6-35B-A3B (hybrid GDN) 536 tok/s @ c16 · 472 tok/s @ c12 · 262K context · image+video · tools + reasoning parsers · full CUDA-graph capture 🐳 docker pull

Hot off the press 🔥 AEON vLLM Ultimate → v0.24.0. Link below ⤵️ The ultimate container for AI on NVIDIA DGX Spark — vLLM compiled from source for GB10/sm_121a. 🔓 UNLOCKED (impossible until this release): ▸ DFlash + FP8 KV cache — 2× KV capacity, draft acceptance intact, measured on-device ▸ FP8 KV on vision models — the KV-bandwidth lever is back ▸ DFlash on FlashInfer & Triton backends (was FlashAttention-only) 🛠 Carried ahead of upstream (still-open PRs, merged + validated): ▸ Triton NVFP4 KV cache → up to 3× KV capacity ▸ DFlash sliding-window drafters + prefix-cache stability for DFlash ▸ High-concurrency DFlash — clean through c=64 🧰 Fixes stock v0.24.0 doesn't have: ▸ Gemma-4 tied embeddings (ModelOpt) — stock crashes at load, ours doesn't ▸ Tool calling out of the box (stock --no-deps builds 500) ▸ Unified-memory hardened — no silent KV overcommit on the 121 GB pool 📊 One image, the whole fleet: Gemma-4-26B-A4B · Qwen3.6-27B · Qwen3.6-35B-A3B (hybrid GDN) 536 tok/s @ c16 · 472 tok/s @ c12 · 262K context · image+video · tools + reasoning parsers · full CUDA-graph capture 🐳 docker pull

ÆON FORGE ✨

24,800 views • 1 month ago

I know some bros want to see what happens when we push the value even further, so here we go: ⚪️ -ncmoe 25 → 41.8 tok/s RAM: 12GB & VRAM: 6.9GB 🔴 -ncmoe 23 → 43.8 tok/s RAM: 12.2GB & VRAM: 7.4GB 🟢 -ncmoe 21 → 38.6 tok/s RAM: 12.4GB & VRAM: 7.8GB 🔵 -ncmoe 19 → 19.8 tok/s RAM: 13.8GB & VRAM: 7.8GB As you can see, there's a sweet spot with the VRAM usage. Play around and monitor to find the right value for your setup, you can use llama.cpp web ui to monitor speeds easily Sweet spot seems to be 25-23 for 8GB VRAM ✅ +40tok/s for Qwen3.6 35B with 64k q8_0 context on a 8GB card is very impressive just by using base llama.cpp, and we didn't even try Turboquant, MTP or Dflash yet! I'll focus on these next 👀 (Server flags and setup in the quoted tweet)

I know some bros want to see what happens when we push the value even further, so here we go: ⚪️ -ncmoe 25 → 41.8 tok/s RAM: 12GB & VRAM: 6.9GB 🔴 -ncmoe 23 → 43.8 tok/s RAM: 12.2GB & VRAM: 7.4GB 🟢 -ncmoe 21 → 38.6 tok/s RAM: 12.4GB & VRAM: 7.8GB 🔵 -ncmoe 19 → 19.8 tok/s RAM: 13.8GB & VRAM: 7.8GB As you can see, there's a sweet spot with the VRAM usage. Play around and monitor to find the right value for your setup, you can use llama.cpp web ui to monitor speeds easily Sweet spot seems to be 25-23 for 8GB VRAM ✅ +40tok/s for Qwen3.6 35B with 64k q8_0 context on a 8GB card is very impressive just by using base llama.cpp, and we didn't even try Turboquant, MTP or Dflash yet! I'll focus on these next 👀 (Server flags and setup in the quoted tweet)

left curve dev

18,049 views • 3 months ago

nemotron 3 omni q8 on dgx spark 128gb vram cranking via hermes agent at 56 tok/s. first night of real local agentic on this box and local hits harder than i thought it would. q8 (near lossless quant, perplexity loss <1% vs fp16) running 256k context on 33 gb of unified memory, 90+ gb still free. multimodal omni weights included. hermes agent driving from telegram, talking to it from bed. speed: 56 tok/s generation, 1,300 tok/s prefill. for context, qwen 3.6 27b at q4 (heavy quant) on 3090 = 40 tok/s. nemotron at higher precision quant on spark beats qwen at lower precision quant on 3090. moe 3.5b active params architecture earns its keep. what i tested tonight: agentic tool calling works clean. ask it to check disks, it autonomously runs df -h through hermes agent. ask it to set up telegram gateway, it invokes the hermes-agent skill, walks through the prompts, completes the flow. overthinks a bit before tool calls (reasoning model trait) but lands the right move every time. researches api docs, internalizes, tests, documents. completes tasks. current models on dgx spark: 9 gguf files, 305 gb total, mix of qwen 3.6 27b dense (5 quants), nemotron omni (4 quants), deepseek v4-flash 158b q4 (the 112gb flagship test). more data coming this week as i benchmark each.

nemotron 3 omni q8 on dgx spark 128gb vram cranking via hermes agent at 56 tok/s. first night of real local agentic on this box and local hits harder than i thought it would. q8 (near lossless quant, perplexity loss <1% vs fp16) running 256k context on 33 gb of unified memory, 90+ gb still free. multimodal omni weights included. hermes agent driving from telegram, talking to it from bed. speed: 56 tok/s generation, 1,300 tok/s prefill. for context, qwen 3.6 27b at q4 (heavy quant) on 3090 = 40 tok/s. nemotron at higher precision quant on spark beats qwen at lower precision quant on 3090. moe 3.5b active params architecture earns its keep. what i tested tonight: agentic tool calling works clean. ask it to check disks, it autonomously runs df -h through hermes agent. ask it to set up telegram gateway, it invokes the hermes-agent skill, walks through the prompts, completes the flow. overthinks a bit before tool calls (reasoning model trait) but lands the right move every time. researches api docs, internalizes, tests, documents. completes tasks. current models on dgx spark: 9 gguf files, 305 gb total, mix of qwen 3.6 27b dense (5 quants), nemotron omni (4 quants), deepseek v4-flash 158b q4 (the 112gb flagship test). more data coming this week as i benchmark each.

Sudo su

30,228 views • 3 months ago

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 views • 3 months ago

nvidia's 3B mamba destroyed alibaba's 3B deltanet on the same RTX 3090. only 24 days between releases. same active parameters, same VRAM tier, completely different architectures. nemotron cascade 2: 187 tok/s. flat from 4K to 625K context. zero speed loss. flags: -ngl 99 -np 1. that's it. no context flags, no KV cache tricks. auto-allocates 625K. qwen 3.5 35B-A3B: 112 tok/s. flat from 4K to 262K context. zero speed loss. flags: -ngl 99 -np 1 -c 262144 --cache-type-k q8_0 --cache-type-v q8_0. needed KV cache quantization to fit 262K. both models held a flat line across every context level. both architectures are context-independent. but nvidia's mamba2 is 67% faster at generating tokens on the exact same hardware and needs fewer flags to get there. same node, same GPU, same everything. the only variable is the model. gold medal math olympiad winner running at 187 tokens per second on single RTX 3090 a card from 6 years ago. nvidia cooked.

nvidia's 3B mamba destroyed alibaba's 3B deltanet on the same RTX 3090. only 24 days between releases. same active parameters, same VRAM tier, completely different architectures. nemotron cascade 2: 187 tok/s. flat from 4K to 625K context. zero speed loss. flags: -ngl 99 -np 1. that's it. no context flags, no KV cache tricks. auto-allocates 625K. qwen 3.5 35B-A3B: 112 tok/s. flat from 4K to 262K context. zero speed loss. flags: -ngl 99 -np 1 -c 262144 --cache-type-k q8_0 --cache-type-v q8_0. needed KV cache quantization to fit 262K. both models held a flat line across every context level. both architectures are context-independent. but nvidia's mamba2 is 67% faster at generating tokens on the exact same hardware and needs fewer flags to get there. same node, same GPU, same everything. the only variable is the model. gold medal math olympiad winner running at 187 tokens per second on single RTX 3090 a card from 6 years ago. nvidia cooked.

Sudo su

187,008 views • 4 months ago

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

401,804 views • 3 months ago

Laguna S 2.1 is built to run locally. We heard the ask for a serious open model in the 80–120B range. Laguna S 2.1 is our answer, and we want it to redefine what this size class can do. Weights ship in BF16, FP8, INT4 and NVFP4, with official GGUF and MLX conversions, plus official DFlash draft models for faster inference. Day-one support across llama.cpp, ollama, Valerie Lee Make Up, SGLang, NVIDIA TRT-LLM, Hugging Face transformers and ZML. Watch it run on an NVIDIA AI DGX Spark.

Laguna S 2.1 is built to run locally. We heard the ask for a serious open model in the 80–120B range. Laguna S 2.1 is our answer, and we want it to redefine what this size class can do. Weights ship in BF16, FP8, INT4 and NVFP4, with official GGUF and MLX conversions, plus official DFlash draft models for faster inference. Day-one support across llama.cpp, ollama, Valerie Lee Make Up, SGLang, NVIDIA TRT-LLM, Hugging Face transformers and ZML. Watch it run on an NVIDIA AI DGX Spark.

Poolside

31,242 views • 24 days ago

$I nearly 2x'd the speed while only using +1GB VRAM with the new MTP update in llama.cpp 🤯 You need to add these flags to start using it: --spec-type draft-mtp \ --spec-draft-p-min 0.75 \ --spec-draft-n-max 2 My results with Qwen3.6 27B on a single RTX 5080 ↓ ⚪️ no flag (without mtp) → 54.3 tok/s with 13.26GB VRAM 🔵 --spec-draft-n-max 2 → 90.7 tok/s with 14.29GB VRAM 🔴 --spec-draft-n-max 2 --spec-draft-p-min 0.75 → 93.9 tok/s with 14.30GB VRAM 🟢 --spec-draft-n-max 6 --spec-draft-p-min 0.75 → 93.9 tok/s with 14.87GB VRAM Increasing to 6 draft tokens didn't help my setup for some reason. I made sure to test with a low context length to have enough headroom and eliminate risk of vram stress. From my understanding: 1) The speed gains are very task-dependent. You need to test across a wide range of tasks to get a realistic idea of the benefits 2) We’re already running heavily quantized GGUF models (Q3, Q4, Q6, etc.), so we already benefit from strong speed/performance thanks to the reduced size. That’s why some people are seeing little to no improvement compared to MLX or other quantized versions The progress over the past few days has been insane to say the least. However, MTP now consumes significantly more VRAM. Personally 16GB just isn't enough to use MTP and run it with a good context size. Time to upgrade lads, 24GB+ users are eating GOOD today 🔥 Full setup below ↓$

I nearly 2x'd the speed while only using +1GB VRAM with the new MTP update in llama.cpp 🤯 You need to add these flags to start using it: --spec-type draft-mtp \ --spec-draft-p-min 0.75 \ --spec-draft-n-max 2 My results with Qwen3.6 27B on a single RTX 5080 ↓ ⚪️ no flag (without mtp) → 54.3 tok/s with 13.26GB VRAM 🔵 --spec-draft-n-max 2 → 90.7 tok/s with 14.29GB VRAM 🔴 --spec-draft-n-max 2 --spec-draft-p-min 0.75 → 93.9 tok/s with 14.30GB VRAM 🟢 --spec-draft-n-max 6 --spec-draft-p-min 0.75 → 93.9 tok/s with 14.87GB VRAM Increasing to 6 draft tokens didn't help my setup for some reason. I made sure to test with a low context length to have enough headroom and eliminate risk of vram stress. From my understanding: 1) The speed gains are very task-dependent. You need to test across a wide range of tasks to get a realistic idea of the benefits 2) We’re already running heavily quantized GGUF models (Q3, Q4, Q6, etc.), so we already benefit from strong speed/performance thanks to the reduced size. That’s why some people are seeing little to no improvement compared to MLX or other quantized versions The progress over the past few days has been insane to say the least. However, MTP now consumes significantly more VRAM. Personally 16GB just isn't enough to use MTP and run it with a good context size. Time to upgrade lads, 24GB+ users are eating GOOD today 🔥 Full setup below ↓

left curve dev

30,086 views • 3 months ago

$I just crammed the updated Gemma 4 26B A4B QAT (MoE) with 180k context into an 8GB RTX 4060 (8 GB VRAM + 16 GB RAM only!!) and optimized the batch size. 23 tokens/sec decode, 300 tokens/sec prefill Yesterday I showed you a Gemma 4 31B dense model running flawlessly on an RTX 4090. Today, we're breaking the VRAM bank on a budget card using Unsloth’s new Gemma 4 26B (A4B) QAT quants. Following Google’s chat template update that boosted agentic benchmarks by +10%, I pushed this model to its absolute limits. Here is how you squeeze 250k context out of 8GB of VRAM. # The Setup & The Optimization - Hardware: Nvidia RTX 4060 (8GB VRAM) + 16GB System RAM - Environment: CUDA 13.0 build of llama.cpp - Model: gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf - Prompt: 28,000 tokens of prompt for each run If you read my L2 cache breakdown (attached in replies), you know the 4060’s 24MB cache maxes out at `-b 1024 -ub 1024`. Push past that, and prefill crashes. I locked those flags in for every test below to ensure maximum GEMM throughput. # 1. The Raw Context Push (Unquantized KV Cache) First, I wanted to see how far pure 8GB VRAM + 16GB RAM could stretch without touching the KV cache: - 80k Context: Prefill 385 t/s | Decode 25.5 t/s - 120k Context: Prefill 270 t/s | Decode 24 t/s llama.cpp flags: .\llama-server -m gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf -c 120000 --port 8080 -ub 1024 -b 1024 Without KV quantization, 120k is your hard ceiling. push past that prefill throughput drops off a cliff, making the model practically unusable for large agentic workloads. # 2. The Q8 KV Cache Lifeline To survive 250k context on a budget card, you have to quantize the KV cache. I enabled 8 bit KV cache (`-ctk q8_0 -ctv q8_0`) and re ran: - 180k Context: Prefill 280 t/s | Decode 22.8 t/s - 250k Context: Prefill 115 t/s | Decode 20 t/s llama.cpp flags: .\llama-server -m gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf -c 180000 --port 8080 -b 1024 -ub 1024 -ctk q8_0 -ctv q8_0 Result: Q8 KV cache brings 250k context back from the dead. Decode speed stabilizes at a highly usable 20 t/s. You are trading a very small bit amount of reasoning precision for an extra 130,000 tokens of context window. if you own a single rtx 3050, 3060, 3070, 4050, 4060, 5050 or 5060, you must try this model and optimize your batch size for higher prefill. Hugging Face links to the updated Unsloth's QAT quants and performance graph are in the replies below. What model are you running on your 6GB, 8GB or 12GB cards right now? Let's see your setups.$

I just crammed the updated Gemma 4 26B A4B QAT (MoE) with 180k context into an 8GB RTX 4060 (8 GB VRAM + 16 GB RAM only!!) and optimized the batch size. 23 tokens/sec decode, 300 tokens/sec prefill Yesterday I showed you a Gemma 4 31B dense model running flawlessly on an RTX 4090. Today, we're breaking the VRAM bank on a budget card using Unsloth’s new Gemma 4 26B (A4B) QAT quants. Following Google’s chat template update that boosted agentic benchmarks by +10%, I pushed this model to its absolute limits. Here is how you squeeze 250k context out of 8GB of VRAM. # The Setup & The Optimization - Hardware: Nvidia RTX 4060 (8GB VRAM) + 16GB System RAM - Environment: CUDA 13.0 build of llama.cpp - Model: gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf - Prompt: 28,000 tokens of prompt for each run If you read my L2 cache breakdown (attached in replies), you know the 4060’s 24MB cache maxes out at `-b 1024 -ub 1024`. Push past that, and prefill crashes. I locked those flags in for every test below to ensure maximum GEMM throughput. # 1. The Raw Context Push (Unquantized KV Cache) First, I wanted to see how far pure 8GB VRAM + 16GB RAM could stretch without touching the KV cache: - 80k Context: Prefill 385 t/s | Decode 25.5 t/s - 120k Context: Prefill 270 t/s | Decode 24 t/s llama.cpp flags: .\llama-server -m gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf -c 120000 --port 8080 -ub 1024 -b 1024 Without KV quantization, 120k is your hard ceiling. push past that prefill throughput drops off a cliff, making the model practically unusable for large agentic workloads. # 2. The Q8 KV Cache Lifeline To survive 250k context on a budget card, you have to quantize the KV cache. I enabled 8 bit KV cache (`-ctk q8_0 -ctv q8_0`) and re ran: - 180k Context: Prefill 280 t/s | Decode 22.8 t/s - 250k Context: Prefill 115 t/s | Decode 20 t/s llama.cpp flags: .\llama-server -m gemma-4-26B-A4B-it-qat-UD-Q4_K_XL.gguf -c 180000 --port 8080 -b 1024 -ub 1024 -ctk q8_0 -ctv q8_0 Result: Q8 KV cache brings 250k context back from the dead. Decode speed stabilizes at a highly usable 20 t/s. You are trading a very small bit amount of reasoning precision for an extra 130,000 tokens of context window. if you own a single rtx 3050, 3060, 3070, 4050, 4060, 5050 or 5060, you must try this model and optimize your batch size for higher prefill. Hugging Face links to the updated Unsloth's QAT quants and performance graph are in the replies below. What model are you running on your 6GB, 8GB or 12GB cards right now? Let's see your setups.

Alok

36,617 views • 22 days ago

Anyone with 8GB or 12GB VRAM setups needs to understand that "-ncmoe" is the key flag to boost performance on llama.cpp Here are my results for Qwen3.6 35B A3B, with 64k q8_0 context on a 8GB RTX 3070Ti: ⚪️ no flag → 8.7 tok/s RAM: 13.6GB & VRAM: 7.8GB 🔴 -ncmoe 35 → 27.5 tok/s RAM: 12.1GB & VRAM: 4.3GB 🟢 -ncmoe 30 → 32.5 tok/s RAM: 12GB & VRAM: 5.6GB 🔵 -ncmoe 25 → 40.9 tok/s RAM: 12GB & VRAM: 6.9GB Please note the ram and vram usage you see are total usage of a windows pc, with the model running. My friend's setup: 8GB VRAM and 16GB RAM. You can boost performance by switching to Linux, just something to keep in mind. Basically, this flag keeps the MoE experts in the first X layers on your CPU + RAM, instead of eating all your VRAM straight away. This is a smart hybrid offload way that lets you run bigger models without OOM while keeping the rest on your GPU for speed. As we can see on the data, there's a sweet spot. When we lower it from 35 to 25, speed bumps +50% because there are more layers on your GPU (look at the VRAM usage). The key here is to play around with the number and fit as much as possible on your VRAM, goal is to have 1GB/800MB headroom to avoid stress. ↓ server flags below

Anyone with 8GB or 12GB VRAM setups needs to understand that "-ncmoe" is the key flag to boost performance on llama.cpp Here are my results for Qwen3.6 35B A3B, with 64k q8_0 context on a 8GB RTX 3070Ti: ⚪️ no flag → 8.7 tok/s RAM: 13.6GB & VRAM: 7.8GB 🔴 -ncmoe 35 → 27.5 tok/s RAM: 12.1GB & VRAM: 4.3GB 🟢 -ncmoe 30 → 32.5 tok/s RAM: 12GB & VRAM: 5.6GB 🔵 -ncmoe 25 → 40.9 tok/s RAM: 12GB & VRAM: 6.9GB Please note the ram and vram usage you see are total usage of a windows pc, with the model running. My friend's setup: 8GB VRAM and 16GB RAM. You can boost performance by switching to Linux, just something to keep in mind. Basically, this flag keeps the MoE experts in the first X layers on your CPU + RAM, instead of eating all your VRAM straight away. This is a smart hybrid offload way that lets you run bigger models without OOM while keeping the rest on your GPU for speed. As we can see on the data, there's a sweet spot. When we lower it from 35 to 25, speed bumps +50% because there are more layers on your GPU (look at the VRAM usage). The key here is to play around with the number and fit as much as possible on your VRAM, goal is to have 1GB/800MB headroom to avoid stress. ↓ server flags below

left curve dev

166,478 views • 3 months ago

first impressions of qwen 3.5 27B dense on a single RTX 3090. 35 tok/s. from 4K all the way to 300K+ context. no speed drop. hermes 4.3 started at 35 and degraded to 15 as context filled. qwen dense holds. MoE held 112 flat. 3x faster but only 3B of 35B active per token. architecture tradeoff. Q4_K_M on 16.7GB. native context 262K. pushed past training limit to 376K before VRAM ceiling on 24GB. tried q8 KV cache at 262K, speed collapsed to 11 tok/s. q4_0 KV is the sweet spot. flash attention mandatory. built in reasoning mode. the model thinks step by step before it answers. full chain of thought surviving Q4 quant. 1,799+ token thinking chains with self correction loops. on a single consumer GPU. gave it one prompt: "build a realtime particle galaxy simulation in one HTML file." 3,340 tokens. 95 seconds. one shot. ran on first load. full reasoning and coding in the video below. optimal config if you want to skip the hours of testing: llama-server -ngl 99 -c 262144 -fa on --cache-type-k q4_0 --cache-type-v q4_0 this is just the warmup. octopus invaders is next: 10 files, 3,400+ lines, zero steering. the prompt hermes quit at 22%. already more impressed than expected. full results coming soon.

first impressions of qwen 3.5 27B dense on a single RTX 3090. 35 tok/s. from 4K all the way to 300K+ context. no speed drop. hermes 4.3 started at 35 and degraded to 15 as context filled. qwen dense holds. MoE held 112 flat. 3x faster but only 3B of 35B active per token. architecture tradeoff. Q4_K_M on 16.7GB. native context 262K. pushed past training limit to 376K before VRAM ceiling on 24GB. tried q8 KV cache at 262K, speed collapsed to 11 tok/s. q4_0 KV is the sweet spot. flash attention mandatory. built in reasoning mode. the model thinks step by step before it answers. full chain of thought surviving Q4 quant. 1,799+ token thinking chains with self correction loops. on a single consumer GPU. gave it one prompt: "build a realtime particle galaxy simulation in one HTML file." 3,340 tokens. 95 seconds. one shot. ran on first load. full reasoning and coding in the video below. optimal config if you want to skip the hours of testing: llama-server -ngl 99 -c 262144 -fa on --cache-type-k q4_0 --cache-type-v q4_0 this is just the warmup. octopus invaders is next: 10 files, 3,400+ lines, zero steering. the prompt hermes quit at 22%. already more impressed than expected. full results coming soon.

Sudo su

120,788 views • 5 months ago

testing Qwen3.5-35B-A3B latest optimized version by UnslothAI on a single RTX 3090. one detailed prompt. zero handholding. watch a 3B model scaffold an entire multifile game project autonomously. the setup: > model: Qwen3.5-35B-A3B (80B total, only 3B active per token) > quant: UD-Q4_K_XL by Unsloth (MXFP4 layers removed in latest update) > speed: 112 tok/s generation, ~130 tok/s prefill > context: 262K tokens > flags: -ngl 99 -c 262144 -np 1 --cache-type-k q8_0 --cache-type-v q8_0 > engine: llama.cpp > agent: Claude Code talk to localhost:8080 (llama.cpp now has native Anthropic API endpoint. no LiteLLM needed) q8_0 KV cache cuts VRAM usage in half vs f16 at 262K. -np 1 is default but worth noting. parallel slots multiply KV cache and at 262K that's an instant OOM. the prompt was more detailed than this but you get the idea: build a space shooter with parallax backgrounds, particle systems, procedural audio, 4 enemy types, boss fights, power-up system, and ship upgrades. 8 JavaScript modules. no libraries. game's called Octopus Invaders. gameplay footage dropping next.

testing Qwen3.5-35B-A3B latest optimized version by UnslothAI on a single RTX 3090. one detailed prompt. zero handholding. watch a 3B model scaffold an entire multifile game project autonomously. the setup: > model: Qwen3.5-35B-A3B (80B total, only 3B active per token) > quant: UD-Q4_K_XL by Unsloth (MXFP4 layers removed in latest update) > speed: 112 tok/s generation, ~130 tok/s prefill > context: 262K tokens > flags: -ngl 99 -c 262144 -np 1 --cache-type-k q8_0 --cache-type-v q8_0 > engine: llama.cpp > agent: Claude Code talk to localhost:8080 (llama.cpp now has native Anthropic API endpoint. no LiteLLM needed) q8_0 KV cache cuts VRAM usage in half vs f16 at 262K. -np 1 is default but worth noting. parallel slots multiply KV cache and at 262K that's an instant OOM. the prompt was more detailed than this but you get the idea: build a space shooter with parallax backgrounds, particle systems, procedural audio, 4 enemy types, boss fights, power-up system, and ship upgrades. 8 JavaScript modules. no libraries. game's called Octopus Invaders. gameplay footage dropping next.

Sudo su

167,244 views • 5 months ago

MiniMax M2.7 is 230B params. Can you actually run it at home? I tested Unsloth's UD-IQ3_XXS (80GB) on 4 different rigs: 🟠 4x RTX 4090 (96GB): 71.52 tok/s, TTFT 1045ms 🟢 4x RTX 5090 (128GB): 120.54 tok/s, TTFT 725ms 🟡 1x RTX PRO 6000 (96GB): 118.74 tok/s, TTFT 765ms 🟣 DGX Spark (128GB) — 24.41 tok/s, TTFT 741ms Backend: llama.cpp. Context: 32k. Max tokens: 4096. I went with IQ3_XXS because it's the biggest quant that fits in 96GB VRAM while still leaving safe headroom for 32k context. Same quant across all four rigs, fairest comparison I could run. Now look at rough peak GPU power draw: 🟠 4x4090 → 1,800W peak (450W × 4) 🟢 4x5090 → 2,300W peak (575W × 4) 🟡 RTX PRO 6000 → 600W peak 🟣 DGX Spark → 240W peak (whole system) The RTX PRO 6000 is the quiet winner. One card, 96GB, matching a 4x5090 rig at roughly a quarter of the power and zero multi-GPU headaches. Best tokens-per-watt by a wide margin. DGX Spark is slow on generation but pulls the least power of any rig here, around 240W for the whole system. Prefill-friendly, memory-rich, wall-socket-friendly. And yes, plenty of people cap their cards. Even then, 4x 4090 or 4x 5090 still pulls well over 1,200W from the GPUs alone.

MiniMax M2.7 is 230B params. Can you actually run it at home? I tested Unsloth's UD-IQ3_XXS (80GB) on 4 different rigs: 🟠 4x RTX 4090 (96GB): 71.52 tok/s, TTFT 1045ms 🟢 4x RTX 5090 (128GB): 120.54 tok/s, TTFT 725ms 🟡 1x RTX PRO 6000 (96GB): 118.74 tok/s, TTFT 765ms 🟣 DGX Spark (128GB) — 24.41 tok/s, TTFT 741ms Backend: llama.cpp. Context: 32k. Max tokens: 4096. I went with IQ3_XXS because it's the biggest quant that fits in 96GB VRAM while still leaving safe headroom for 32k context. Same quant across all four rigs, fairest comparison I could run. Now look at rough peak GPU power draw: 🟠 4x4090 → 1,800W peak (450W × 4) 🟢 4x5090 → 2,300W peak (575W × 4) 🟡 RTX PRO 6000 → 600W peak 🟣 DGX Spark → 240W peak (whole system) The RTX PRO 6000 is the quiet winner. One card, 96GB, matching a 4x5090 rig at roughly a quarter of the power and zero multi-GPU headaches. Best tokens-per-watt by a wide margin. DGX Spark is slow on generation but pulls the least power of any rig here, around 240W for the whole system. Prefill-friendly, memory-rich, wall-socket-friendly. And yes, plenty of people cap their cards. Even then, 4x 4090 or 4x 5090 still pulls well over 1,200W from the GPUs alone.

stevibe

191,782 views • 3 months ago

THIS AMERICAN DEVELOPER SPENT WEEKS DEBUGGING TIMEOUT ERRORS IN OLLAMA. THEN HE LOOKED UNDER THE HOOD LM Studio is just llama.cpp Ollama is just llama.cpp so he cloned llama.cpp from source, pulled Qwen 3.6 35B off Hugging Face, set up asymmetric KV quantization and got a local server running on 127.0.0.1:8080 plugged it into VS Code, connected it to OpenClaw, 53 tok/s on an M1 Max with 262K context zero wrappers, zero timeout errors, zero API fees bookmark & like this before your next timeout error hits full breakdown of my raw llama.cpp setup ↓

THIS AMERICAN DEVELOPER SPENT WEEKS DEBUGGING TIMEOUT ERRORS IN OLLAMA. THEN HE LOOKED UNDER THE HOOD LM Studio is just llama.cpp Ollama is just llama.cpp so he cloned llama.cpp from source, pulled Qwen 3.6 35B off Hugging Face, set up asymmetric KV quantization and got a local server running on 127.0.0.1:8080 plugged it into VS Code, connected it to OpenClaw, 53 tok/s on an M1 Max with 262K context zero wrappers, zero timeout errors, zero API fees bookmark & like this before your next timeout error hits full breakdown of my raw llama.cpp setup ↓

leopardracer

238,950 views • 3 months ago

We just released Photoshop AI Assistant 🎉 I edited my images with text keeping consistency. It understands context of the image, so can select particular parts of the image you want to edit. And, of course, it can name your layers 😅 and do many other things. Enjoy!

We just released Photoshop AI Assistant 🎉 I edited my images with text keeping consistency. It understands context of the image, so can select particular parts of the image you want to edit. And, of course, it can name your layers 😅 and do many other things. Enjoy!

Kris Kashtanova

18,097 views • 1 month ago

Unsloth AI and NVIDIA are Revolutionizing Local LLM Fine-Tuning: From RTX Desktops to DGX Spark Fine-tune popular AI models faster with Unsloth on NVIDIA RTX AI PCs such as GeForce RTX desktops and laptops to RTX PRO workstations and the new DGX Spark to build personalized assistants for coding, creative work, and complex agentic workflows. The landscape of modern AI is shifting. We are moving away from a total reliance on massive, generalized cloud models and entering the era of local, agentic AI. Whether it is tuning a chatbot to handle hyper-specific product support or building a personal assistant that manages intricate schedules, the potential for generative AI on local hardware is boundless. However, developers face a persistent bottleneck: How do you get a Small Language Model (SLM) to punch above its weight class and respond with high accuracy for specialized tasks? The answer is Fine-Tuning, and the tool of choice is Unsloth. Unsloth provides an easy and high-speed method to customize models. Optimized for efficient, low-memory training on NVIDIA GPUs, Unsloth scales effortlessly from GeForce RTX desktops and laptop all the way to the DGX Spark, the world’s smallest AI supercomputer...... Full analysis: NVIDIA NVIDIA AI NVIDIA AIDev NVIDIAnewsroom Unsloth AI Unsloth

Unsloth AI and NVIDIA are Revolutionizing Local LLM Fine-Tuning: From RTX Desktops to DGX Spark Fine-tune popular AI models faster with Unsloth on NVIDIA RTX AI PCs such as GeForce RTX desktops and laptops to RTX PRO workstations and the new DGX Spark to build personalized assistants for coding, creative work, and complex agentic workflows. The landscape of modern AI is shifting. We are moving away from a total reliance on massive, generalized cloud models and entering the era of local, agentic AI. Whether it is tuning a chatbot to handle hyper-specific product support or building a personal assistant that manages intricate schedules, the potential for generative AI on local hardware is boundless. However, developers face a persistent bottleneck: How do you get a Small Language Model (SLM) to punch above its weight class and respond with high accuracy for specialized tasks? The answer is Fine-Tuning, and the tool of choice is Unsloth. Unsloth provides an easy and high-speed method to customize models. Optimized for efficient, low-memory training on NVIDIA GPUs, Unsloth scales effortlessly from GeForce RTX desktops and laptop all the way to the DGX Spark, the world’s smallest AI supercomputer...... Full analysis: NVIDIA NVIDIA AI NVIDIA AIDev NVIDIAnewsroom Unsloth AI Unsloth

Marktechpost AI Dev News ⚡

31,551 views • 7 months ago

HERMES AGENT CROSSED 140,000 GITHUB STARS IN 3 MONTHS AND JUST BECAME THE MOST USED AGENT IN THE WORLD. Most AI agents forget everything between sessions. Hermes writes its own skills from experience. Next time it runs the skill, improves it, and gets faster. Independent benchmarks show agents with 20+ self-created skills complete similar tasks 40% faster than fresh instances. Qwen 3.6 where the 35B version outperforms last year's 120B models at one third the memory footprint. DGX Spark with 128GB unified memory running everything locally at $0 per month after hardware. The setup takes 30 minutes. LM Studio plus Qwen 3.6 27B for the model server. One install script for Hermes. One config connecting them. Set context window to 65,536 tokens or nothing works. After one month of daily use your skills directory has 20 to 50 learned workflows. Your Hermes is genuinely different from anyone else's.

HERMES AGENT CROSSED 140,000 GITHUB STARS IN 3 MONTHS AND JUST BECAME THE MOST USED AGENT IN THE WORLD. Most AI agents forget everything between sessions. Hermes writes its own skills from experience. Next time it runs the skill, improves it, and gets faster. Independent benchmarks show agents with 20+ self-created skills complete similar tasks 40% faster than fresh instances. Qwen 3.6 where the 35B version outperforms last year's 120B models at one third the memory footprint. DGX Spark with 128GB unified memory running everything locally at $0 per month after hardware. The setup takes 30 minutes. LM Studio plus Qwen 3.6 27B for the model server. One install script for Hermes. One config connecting them. Set context window to 65,536 tokens or nothing works. After one month of daily use your skills directory has 20 to 50 learned workflows. Your Hermes is genuinely different from anyone else's.

Lummox

69,676 views • 2 months ago