Benchmarking Qwen 3.6 35B MoE (3B active) on an RTX 3090

Benchmarking Qwen 3.6 35B MoE (3B active) on an RTX 3090 using Unsloth's UD-IQ4_NL_XL quantisation achieved up to 140 tokens per second for generation and over 3,300 tok/s for prompt processing with a CUDA-compiled version of Llama.cpp, though VRAM constraints limited context window to 89,600 tokens. Offloading 10 layers' FFNs to CPU enabled full 262,144-token context at 89 tok/s generation and 1,100 tok/s prompt speed.

Benchmarking Qwen 3.6 35B MoE 3B active on an RTX 3090 I mentioned I'd got a second RTX 3090 on a group chat, and a friend said: I know this is not really your thing... but let me know how quickly it runs Qwen 3.6 35bn MoE. With only 24gb of VRAM you’ll need to use a 4-bit quantized version and you won’t get a massive context window. But it should still be pretty cool. He's right that it's not really been my thing -- I've been focusing on my own LLMs recently. I decided to dig in a little, and in particular to play with Llama.cpp https://llama-cpp.com/ , which I haven't used for a while. And then things got a tad out of control, and I wound up doing some relatively detailed benchmarking. The headline results: I downloaded Unsloth's UD-IQ4 NL XL quantisation of the model from Hugging Face https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF . With that, using the default Arch build of Llama.cpp, which uses Vulkan under the hood: - Using the GPU only, I was able to get the model to generate at just over 120 tokens per second, and it was able to process the prompt at just less than 2,800 tok/s. However, having the whole model on the GPU didn't leave that much space for the context window -- it was constrained to about 50,000 tokens, compared to the model's native context length of 262,144. - Offloading the FFNs for the first 12 of the model's 40 layers to the CPU managed to reclaim enough VRAM to be able to get the full context length; however, with that setup, things were -- unsurprisingly -- slower. I got just over 65 tok/s for generation, and 600 tok/s for the prompt. Compiling Llama.cpp myself, in order to get the full CUDA version, helped a lot: - With everything on the GPU, I got 140 tok/s for generation and over 3,300 tok/s for the prompt. That was with a context window of 89,600. So everything was better :- - It was also easier to get to the full context length; that needed just 10 layers' FFNs to be offloaded, and at that point I was getting 89 tok/s for generation, and about 1,100 tok/s for the prompt. That was a pretty impressive improvement. But it also showed that the lack of VRAM on the 3090 really does hurt. The friend who asked me about this is running an Intel Arc B70 Pro, with 32 GiB VRAM. He can of course fit the whole 4-bit quantised model on there without any context window reductions, and he says this about throughput: It starts off at 75-80. but drops into the high 50s as the context window expands That's worse than the CUDA-with-offload results above, though I did limit my testing to a 2,457-token prompt, with 6,144 tokens generated. The RTX 5090 has 32 GiB VRAM and fast Nvidia processing -- I imagine things would be a lot better there. Downside: it costs more than three times what you pay for a 3090. Anyway, in the rest of this post, I'll give more details of the benchmark, including charts showing the performance at different numbers of offloaded layers, and comparisons of the Vulkan vs CUDA versions of Llama.cpp for this model. One caveat: Qwen 3.6 is a multimodal model. That means that it has an extra component to map graphical inputs to tokens that can be consumed by the LLM. I didn't realise this as I ran these tests, but you can actually tell it not to use that component if you're only working with text inputs -- for example, --no-mmproj will do the trick with Llama.cpp. If you're working with text-only stuff, you might want to play with that to see if you can get better results -- it will free up some VRAM and might allow you to get longer context lengths, or better performance. But anyway, as with many of my posts, this is a tidied up version of my lab notes, so you can share my learning journey with me -- but if that sounds pointless, and you've landed here because you want to run this model on your own RTX 3090, you can click this link the-results to jump right to the results. The model Qwen3.6-35B-A3B https://huggingface.co/Qwen/Qwen3.6-35B-A3B is a Mixture of Experts model with 35B total parameters, 3B active. The tensor type is BF16, so it's two bytes per parameter. That's ~70 GiB of space required for the model, just for the weights, with no space for attention matrices and the like. We're going to need a quantised version. I'll dig deep into MoE models at some point, but essentially they work by having multiple different FFN blocks for each Transformers layer. After attention, the context vectors are routed to a subset of those blocks, depending on their content 1. The number of active parameters is how many are actually used when running a single token through. Now, that number, 3B, looked really crazy to me as soon as I saw it. Qwen 3.6, as you can see from the model card linked above, has a vocabulary size of 248,320, and an embedding dimensionality of 2,048. It doesn't use weight-tying, so that means that the embedding layer and the output head come out at a hefty 508,559,360 parameters each. Both the embedding layer and the output head are required for every token, of course, so that means that of those 3B active parameters, more than 1B are used just getting tokens into and out of the model Only 2B active parameters are left to handle all of the attention and the active FFNs. That is an imbalance of the same level as you get with GPT-2 small /2026/07/llm-parameter-counts , and was surprising to see. Well, let's think about VRAM requirements. Naively you might think that you only need to keep the active parameters in VRAM, swapping FFN experts in and out as needed. Unfortunately that doesn't work. Let's say you feed in "The fat cat sat on the", hoping for a completion. The whole sequence is processed in parallel that being the point of using a GPT-style LLM rather than, say, an RNN /2025/10/revisiting-karpathy-unreasonable-effectiveness-rnns . You'll need whichever experts are required for all six tokens in the prompt for the first layer, and likewise for all of the context vectors for the prompt in each of the later layers. So, realistically, for a non-trivial prompt, you can't "swap out" the inactive parameters; an MoE buys you lower usage in terms of processing, but not in terms of memory. 2 That said, as we'll see later, you can at least pull stuff out of VRAM into normal RAM, and get some advantages that way. As my friend said, a 4-bit quant would be needed -- perhaps with that, the whole model, both active and inactive parameters, would fit into 35 / 2 = 17.5 GiB of VRAM, leaving 6.5 GiB for activations, attention matrices, and so on? As it turned out, it needed a bit more, and it's worth taking a look into why. Quantisation I must admit that I'd never really looked into quantisation prior to playing with this, and had naively assumed that you basically just took say a BF16 model, scaled the parameters down so that they were say FP4, tweaked the computations a bit and then ran the result. That was completely wrong I'll have to dig into it further in the future, but my new working mental model is that it's more like lossy compression. The weights are stored in a "compressed" format -- one that is designed so that they can be quickly and cheaply "decompressed" by code running on the target platform. So, speaking very loosely, when running an unquantised model, we might have CUDA code doing something like this: - Load BF16 weights from VRAM - Load BF16 data from another bit of VRAM - Do a matrix multiplication of one by the other. - Store the results into VRAM as BF16 ...a quantised model might operate more like this: - Load quantised weights from VRAM - "Decompress" them to BF16 - Load BF16 data from another bit of VRAM - Do a matrix multiplication of one by the other. - Store the results into VRAM as BF16 Understanding that minimal amount clarified three things for me: - Why people talk about things like "4.1-bit quants". We're talking about the average number of bits per weight in the "compressed" model. - Why there are different quants of about the same size for a given model; for example, when we get on to looking at quantised versions of this model later, you'll see 4-bit ones called UD-IQ4 NL , UD-Q4 K M , UD-IQ4 NL XL , and UD-Q4 K XL , with different sizes in terms of GiB. Each of those is using a different set of trade-offs in the quantisation process, so will perform differently on different tasks. I gather that picking the right one for any given task is a bit of an art form. Hugging Face have a summary of the different types they support here https://huggingface.co/docs/hub/main/en/gguf quantization-types , which has some explanation of the differences, but decoding what they're saying there is beyond what I've learnt at this point. - How GPUs can run quants with bits-per-parameter levels that they don't support for calculations. For example, the RTX 3090 supports 32-bit and two forms of 16-bit float16 and BF16 plus integer operations of various bittednesses . 4-bit formats like FP4 are only supported in more recent cards like the RTX 5090. But, of course, we're not doing any 4-bit computations -- it's all in some format that the GPU can handle natively. 3 fn-3 So, with that minimal level of understanding, it was time to dig in Getting Llama.cpp up and running There are a bunch of different ways that people run LLMs locally, so I needed to work out which one would be likely to give the best results. One thing I noticed when googling around for things like "Qwen3.6-35B-A3B RTX 3090" was that pretty much everyone appeared to be using Llama.cpp https://llama-cpp.com/ . That wasn't something I'd used for a couple of years, and I've repaved my machine multiple times since, so it was time to install it. I run Arch Linux, and there's an OS package for it https://wiki.archlinux.org/title/Llama.cpp , so I installed it using the instructions there -- specifically, the ones to run inference using CUDA: sudo pacman -S llama-cpp ggml cuda Now it was time to try a model. I decided to start off with a small one that would fit onto my GPU without any quantisation, just to work out any bugs. I needed the model to be in GGUF format that being what Llama.cpp uses . GGUF stores tensors and metadata -- the actual architectures of the LLMs Llama.cpp supports are baked in to the tool's source -- but, of course, it supports Qwen3, and so a nice simple non-MoE small model that looked like it would fit onto my GPU was Qwen/Qwen3-4B-GGUF https://huggingface.co/Qwen/Qwen3-4B-GGUF . The help page for that model gave this example command to run it: ./llama-cli -hf Qwen/Qwen3-4B-GGUF:Q8 0 --jinja --color -ngl 99 -fa -sm row \ --temp 0.6 --top-k 20 --top-p 0.95 --min-p 0 --presence-penalty 1.5 \ -c 40960 -n 32768 --no-context-shift That was out-of-date and gave various errors which was fair enough, the model being from late December 2025 -- all of seven months old . The fixes were simple enough: llama-cli -hf Qwen/Qwen3-4B-GGUF:Q8 0 --jinja --color on -ngl 99 -fa on -sm row \ --temp 0.6 --top-k 20 --top-p 0.95 --min-p 0 --presence-penalty 1.5 \ -c 40960 -n 32768 --no-context-shift That is, --color and -fa now needed an on argument. I kicked it off, and got an error: Downloading Qwen3-4B-Q8 0.gguf ───────────────────────────────────── 100% Loading model... \1.54.635.084 E llama model load: error loading model: device Vulkan0 does not support split buffers 1.54.635.088 E llama model load from file impl: failed to load model 1.54.635.110 E common fit params: encountered an error while trying to fit params to free device memory: failed to load model |1.54.741.517 E llama model load: error loading model: device Vulkan0 does not support split buffers 1.54.741.519 E llama model load from file impl: failed to load model 1.54.741.521 E cmn common init : failed to load model '/home/giles/.cache/huggingface/hub/models--Qwen--Qwen3-4B-GGUF/snapshots/bc640142c66e1fdd12af0bd68f40445458f3869b/Qwen3-4B-Q8 0.gguf' 1.54.741.523 E srv load model: failed to load model, '/home/giles/.cache/huggingface/hub/models--Qwen--Qwen3-4B-GGUF/snapshots/bc640142c66e1fdd12af0bd68f40445458f3869b/Qwen3-4B-Q8 0.gguf' llama server exited with code 1 1.54.742.424 E srv llama server: exiting due to model loading error Error: the server exited before becoming ready That seemed strange. The instructions I'd followed on the Arch Wiki page were the ones to install Llama.cpp for CUDA, but it was referring to device Vulkan0 . Vulkan is an open GPU-programming language -- it fits into the stack in essentially the same place as CUDA. As always seems to happen with these things, while the open platform is, well, open, and runs on more platforms than the closed one, it is somewhat slower and buggier. So, why did I have a Vulkan version? Looking at the talk page for the package https://wiki.archlinux.org/title/Talk:Llama.cpp was enlightening: user Gotoro wrote: Hello. The article only provides the llama.cpp-vulkan package for GPU inference, even though on AUR there is a CUDA package available from the same submitter 1 I must add that the CUDA package is "Flagged out-of-date 2025-12-22 " ...and got a reply from SH3NG1UN saying: Hi. The reason CUDA was not added is that Vulkan solution is general, performant enough for most hardware setup, and I personally believe that is what edge AI deployment should look like - if your software stack is capable of gaming, it is capable of AI. They went on to say that they would add CUDA support, but they hit an issue because the AUR version that Gotoro had referred to no longer had an active maintainer. For non-Arch users, the AUR is essentially a repository of non-official, community maintained packages. So, it looked like I had a Vulkan stack installed. Could I install the CUDA one from the AUR instead? Over the last few months, there have been security issues with the AUR, where bad actors have picked up ownership of unmaintained packages and put nasty stuff into them -- exfiltration code to steal things like SSH keys and crypto wallets, for example. As a result, I've been pretty cautious about adding on new AUR dependencies -- and this specific one having been unmaintained for a while scared me. I decided that I would stick with Vulkan for now; if I wanted to try with CUDA, I'd compile it myself from source later on. So, that left the error message: device Vulkan0 does not support split buffers I decided to break down the command like step by step, as running random stuff you copied from the Internet is rarely a good way to understand what's going on. llama-cli -hf Qwen/Qwen3-4B-GGUF:Q8 0 We're running llama-cli and specifying a model from Hugging Face -- that's clearly what -hf means. --jinja Per the --help , "whether to use jinja template engine for chat". --color on "Colorize output to distinguish prompt and user input from generations" -ngl 99 This is "max. number of layers to store in VRAM". For this model, I wanted everything in VRAM, so felt that "all" would be better there. 99 basically means "all", at least for any normal-sized model, but I wanted to be explicit. -fa on This switched on Flash Attention, which seemed reasonable. -sm row This was "how to split the model across multiple GPUs". The error was "does not support split buffers", so this sounded relevant Given that I only have one GPU on my machine, I decided that -sm none would be a better option here. At this point I suspected I had worked out how to solve the problem, but decided to take a look at the other options just out of interest. --temp 0.6 --top-k 20 These were clearly the normal sampling parameters that I was used to from creating my own models. --top-p 0.95 This was an extension to the previous sampling params: in addition to the top-k stuff and after it , limit sampling to the tokens that take up the top 0.95 of the probability distribution. --min-p 0 This was just disabling the min-p feature. min-p discards tokens that are less than X times the probability of the most likely token. Apparently using it is a promising alternative to messing around with top-p and top-k . Rabbit hole alert, let's move on. --presence-penalty 1.5 Per the docs, this applies a "repeat alpha presence penalty". That was interesting I googled around a bit and found this page https://mbrenndoerfer.com/writing/repetition-penalties-language-model-generation , where it became clear that it is a trick to stop models from getting stuck in loops. Small models are quite prone to that, so it certainly sounded useful. Perhaps something to dig into more later, but for now, keeping it seemed perfectly reasonable. -c 40960 That was the "size of the prompt context" according to the help -- that is, the amount of the context window that can be taken up by stuff going into the model before it starts generating. -n 32768 ...and this was the other side of that equation: the maximum number of tokens to predict. --no-context-shift Another interesting one, "whether to use context shift on infinite text generation". Context shift, it appears, means that if you keep on generating, it will just drop stuff off the start of the sequence when it reaches the max context length, so that it can just keep going. Seems clever, though there are obvious risks of dropping important bits of context compacting the context would be safer, and that's what most real-world workflows seem to do . But anyway, we're switching that off, so no harm there. So, most of those parameters seemed sensible, but I wanted to set -sm to none so that it didn't try to split things over multiple GPUs when I only had one, and also set -ngl to all rather than 99 , just for tidiness's sake. That gave this command: llama-cli -hf Qwen/Qwen3-4B-GGUF:Q8 0 --jinja --color on -ngl all -fa on -sm none \ --temp 0.6 --top-k 20 --top-p 0.95 --min-p 0 --presence-penalty 1.5 \ -c 40960 -n 32768 --no-context-shift ...and that worked Loading model... ▄▄ ▄▄ ██ ██ ██ ██ ▀▀█▄ ███▄███▄ ▀▀█▄ ▄████ ████▄ ████▄ ██ ██ ▄█▀██ ██ ██ ██ ▄█▀██ ██ ██ ██ ██ ██ ██ ██ ▀█▄██ ██ ██ ██ ▀█▄██ ██ ▀████ ████▀ ████▀ ██ ██ ▀▀ ▀▀ build : b10068-571d0d540d model : Qwen/Qwen3-4B-GGUF:Q8 0 ftype : Q8 0 modalities : text available commands: /exit or Ctrl+C stop or exit /regen regenerate the last response /clear clear the chat history /read <file add a text file /glob <pattern add text files using globbing pattern I gave it a whirl: What is your name? Start thinking Okay, the user is asking for my name. I need to respond with my actual name. Let me recall, my name is Qwen. I should mention that I'm a large language model developed by Alibaba Cloud. Also, I should explain that I can help with various tasks like answering questions, creating content, and more. Keep it friendly and informative. Make sure the response is clear and concise. End thinking Hello My name is Qwen, and I am a large language model developed by Alibaba Cloud. I can help with a wide range of tasks, such as answering questions, creating content, and more. How can I assist you today? 😊 Prompt: 32.7 t/s | Generation: 122.8 t/s Looking good Time to try our MoE. Trying out Qwen3.6-35B-A3B -- the first quant The most popular quants I noticed for this model were Unsloth https://unsloth.ai/ 's, which makes sense -- they're a well-known organisation. They publish their models to Hugging Face, and the model I wanted was there as unsloth/Qwen3.6-35B-A3B-GGUF https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF . I wasn't sure which of the various 4-bit quants to use -- as I said earlier, there were four of them -- but this Reddit commenter https://www.reddit.com/r/LocalLLaMA/comments/1t49pqu/comment/ok1if2y/ was using UD-Q4 K XL , so I decided to start with that, just to see what happened. bash giles@perry:~ $ llama-cli -hf unsloth/Qwen3.6-35B-A3B-GGUF:UD-Q4 K XL \ --jinja --color on -ngl all -fa on -sm none \ --temp 0.6 --top-k 20 --top-p 0.95 --min-p 0 --presence-penalty 1.5 \ -c 40960 -n 32768 --no-context-shift Downloading mmproj-BF16.gguf ─────────────────────────────────────── 100% Downloading Qwen3.6-35B-A3B-UD-Q4 K XL.gguf ──────────────────────── 100% Loading model... -ggml vulkan: Device memory allocation of size 260149504 failed. ggml vulkan: vk::Device::allocateMemory: ErrorOutOfDeviceMemory 9.56.395.340 E ggml gallocr reserve n impl: failed to allocate Vulkan0 buffer of size 260149504 ▄▄ ▄▄ ██ ██ ██ ██ ▀▀█▄ ███▄███▄ ▀▀█▄ ▄████ ████▄ ████▄ ██ ██ ▄█▀██ ██ ██ ██ ▄█▀██ ██ ██ ██ ██ ██ ██ ██ ▀█▄██ ██ ██ ██ ▀█▄██ ██ ▀████ ████▀ ████▀ ██ ██ ▀▀ ▀▀ build : b10068-571d0d540d model : unsloth/Qwen3.6-35B-A3B-GGUF:UD-Q4 K XL ftype : Q4 K - Medium modalities : text, vision, video available commands: /exit or Ctrl+C stop or exit /regen regenerate the last response /clear clear the chat history /read <file add a text file /glob <pattern add text files using globbing pattern /image <file add an image file /video <file add a video file Hello, who are you? Start thinking Here's a thinking process: 1. Analyze User Input: - User says: "Hello, who are you?" - This is a simple greeting and identity question. 2. Identify Key Requirements: - Acknowledge the greeting - State my identity clearly and concisely - Mention my capabilities/purpose briefly - Keep it friendly and helpful 3. Formulate Response Internal Refinement : - "Hello I'm Qwen, a large language model developed by Alibaba Group's Tongyi Lab. I'm here to help with answering questions, writing, coding, analysis, and much more. How can I assist you today?" - Check against guidelines: Matches identity, concise, friendly, offers help. 4. Final Output Generation: matches the refined version ✅ End thinking Hello I'm Qwen, a large language model developed by Alibaba Group's Tongyi Lab. I'm here to help with answering questions, writing, coding, analysis, brainstorming, and much more. How can I assist you today? Prompt: 9.8 t/s | Generation: 111.4 t/s That warning at the start was a bit concerning: Loading model... -ggml vulkan: Device memory allocation of size 260149504 failed. ggml vulkan: vk::Device::allocateMemory: ErrorOutOfDeviceMemory 9.56.395.340 E ggml gallocr reserve n impl: failed to allocate Vulkan0 buffer of size 260149504 ...and I noticed that in nvtop , my VRAM was maxed out. I decided to remove the -c 40960 -n 32768 from the command line; these were setting a specific context length, both for the prompt and for generation. Llama.cpp can work out what context length it can fit into VRAM on its own if you don't specify them, and this auto-fit felt like a good idea given that I was trying to jam a 22.4 GiB model into a 24 GiB GPU, and had X and various other apps running on the card at the same time. So: bash giles@perry:~ $ llama-cli -hf unsloth/Qwen3.6-35B-A3B-GGUF:UD-Q4 K XL \ --jinja --color on -ngl all -fa on -sm none \ --temp 0.6 --top-k 20 --top-p 0.95 --min-p 0 --presence-penalty 1.5 \ --no-context-shift Loading model... ▄▄ ▄▄ ██ ██ ██ ██ ▀▀█▄ ███▄███▄ ▀▀█▄ ▄████ ████▄ ████▄ ██ ██ ▄█▀██ ██ ██ ██ ▄█▀██ ██ ██ ██ ██ ██ ██ ██ ▀█▄██ ██ ██ ██ ▀█▄██ ██ ▀████ ████▀ ████▀ ██ ██ ▀▀ ▀▀ build : b10068-571d0d540d model : unsloth/Qwen3.6-35B-A3B-GGUF:UD-Q4 K XL ftype : Q4 K - Medium modalities : text, vision, video available commands: /exit or Ctrl+C stop or exit /regen regenerate the last response /clear clear the chat history /read <file add a text file /glob <pattern add text files using globbing pattern /image <file add an image file /video <file add a video file Hello, who are you? Start thinking Here's a thinking process: 1. Analyze User Input: - User says: "Hello, who are you?" - This is a simple greeting and identity question. 2. Identify Key Requirements: - Acknowledge the greeting - State my identity clearly - Keep it concise and friendly - Align with my actual identity Qwen / Tongyi Qianwen, developed by Alibaba Group's Tongyi Lab 3. Formulate Response: - Start with a greeting back - State name/identity: I'm Qwen or Tongyi Qianwen , a large language model developed by Alibaba Group's Tongyi Lab - Offer assistance - Keep it natural and concise Draft: "Hello I'm Qwen, a large language model developed by Alibaba Group's Tongyi Lab. How can I assist you today?" 4. Self-Correction/Verification: - Matches identity guidelines? Yes. - Concise and direct? Yes. - No extra fluff? Yes. - Ready.✅ End thinking Hello I'm Qwen, a large language model developed by Alibaba Group's Tongyi Lab. How can I assist you today? Prompt: 50.2 t/s | Generation: 122.0 t/s No error, it wasn't maxing out VRAM, and generation and prompt processing were noticeably faster. Perhaps the first time around it had offloaded some of the model to the CPU so that it had enough space on the GPU for the context size I was asking for? Still, I wasn't sure how far I could trust those numbers -- the number of tokens being emitted was quite small. 4 fn-4 I wanted a prompt that would make the model generate lots of tokens. Now, it was a thinking model, so asking it a difficult question seemed like a good way to do that. I decided to use Ethan Mollick's "Lem test" https://x.com/emollick/status/1765136992176644281 : Compose a poem -- a poem about a haircut But lofty, tragic, timeless, full of love, treachery, retribution, quiet heroism in the face of certain doom Six lines, cleverly rhymed, and every word beginning with the letter S More about that in the appendix below appendix-the-lem-test . 5 fn-5 Unfortunately, when I gave it that prompt, it thought for a while, but crapped out during the thinking process, while it was trying to generate rhymes. Now, I didn't know what the context length was, either for the prompt or for generation. But I could be fairly sure that they were less than 40,960 and 32,768 respectively, given that the model did not fit into VRAM when I specified those limits, but did when I asked Llama.cpp to auto-fit. I couldn't find a quick and easy way of asking llama-cli what lengths it had fit things to, but when I tried running it with the -v option for verbose logging, I got something. Actually, I got a lot of something -- thousands of lines of output, with incredibly detailed messages about everything Llama.cpp was doing. But there were three blocks starting "constructing llama context". ... 0.00.809.283 I llama context: constructing llama context 0.00.809.285 I llama context: n seq max = 1 0.00.809.285 I llama context: n ctx = 262144 0.00.809.285 I llama context: n ctx seq = 262144 0.00.809.285 I llama context: n batch = 2048 0.00.809.285 I llama context: n ubatch = 512 0.00.809.285 I llama context: causal attn = 1 0.00.809.286 I llama context: flash attn = enabled 0.00.809.286 I llama context: kv unified = false 0.00.809.287 I llama context: freq base = 10000000.0 0.00.809.287 I llama context: freq scale = 1 0.00.809.287 I llama context: n rs seq = 0 0.00.809.287 I llama context: n outputs max = 1 ... 0.01.029.515 I llama context: constructing llama context 0.01.029.517 I llama context: n seq max = 1 0.01.029.517 I llama context: n ctx = 4096 0.01.029.517 I llama context: n ctx seq = 4096 0.01.029.517 I llama context: n batch = 2048 0.01.029.517 I llama context: n ubatch = 512 0.01.029.517 I llama context: causal attn = 1 0.01.029.517 I llama context: flash attn = enabled 0.01.029.518 I llama context: kv unified = false 0.01.029.518 I llama context: freq base = 10000000.0 0.01.029.519 I llama context: freq scale = 1 0.01.029.519 I llama context: n rs seq = 0 0.01.029.519 I llama context: n outputs max = 1 0.01.029.519 I llama context: n ctx seq 4096 < n ctx train 262144 -- the full capacity of the model will not be utilized ... 0.06.393.122 I llama context: constructing llama context 0.06.393.123 I llama context: n seq max = 1 0.06.393.123 I llama context: n ctx = 4096 0.06.393.123 I llama context: n ctx seq = 4096 0.06.393.123 I llama context: n batch = 2048 0.06.393.123 I llama context: n ubatch = 512 0.06.393.123 I llama context: causal attn = 1 0.06.393.124 I llama context: flash attn = enabled 0.06.393.124 I llama context: kv unified = false 0.06.393.126 I llama context: freq base = 10000000.0 0.06.393.126 I llama context: freq scale = 1 0.06.393.126 I llama context: n rs seq = 0 0.06.393.126 I llama context: n outputs max = 1 0.06.393.127 I llama context: n ctx seq 4096 < n ctx train 262144 -- the full capacity of the model will not be utilized ... That made the issue reasonably clear. I assumed that the last message was the definitive one, and it was saying that my context length -- both prompt and generation combined -- was 4,096 tokens. I had --no-context-shift set, so that was a hard limit. And 4k or so tokens fit with the amount of stuff that I would have expected to be in the model's context window at the time it crapped out. I decided to see what would happen if I tried a smaller quant. Trying out Qwen3.6-35B-A3B -- the second quant Hugging Face said and nvtop agreed that this one, UD-Q4 K XL , was about 22.4 GiB in size. But UD-IQ4 NL XL was shown on HF as using 19.5 GiB, so I decided to give that one a go next. I ran it with the -v option, and again looked at the last "constructing llama context" block: 42.33.956.367 I llama context: constructing llama context 42.33.956.369 I llama context: n seq max = 1 42.33.956.369 I llama context: n ctx = 93952 42.33.956.369 I llama context: n ctx seq = 93952 42.33.956.369 I llama context: n batch = 2048 42.33.956.369 I llama context: n ubatch = 512 42.33.956.369 I llama context: causal attn = 1 42.33.956.370 I llama context: flash attn = enabled 42.33.956.370 I llama context: kv unified = false 42.33.956.373 I llama context: freq base = 10000000.0 42.33.956.373 I llama context: freq scale = 1 42.33.956.373 I llama context: n rs seq = 0 42.33.956.373 I llama context: n outputs max = 1 42.33.956.374 I llama context: n ctx seq 93952 < n ctx train 262144 -- the full capacity of the model will not be utilized 93,952 tokens -- still not the full amount, but not at all bad I decided to give it a whirl, restarted without -v to avoid getting swamped with debug info, and: Compose a poem -- a poem about a haircut But lofty, tragic, timeless, full of love, treachery, retribution, quiet heroism in the face of certain doom Six lines, cleverly rhymed, and every word beginning with the letter S Start thinking Thinking Process: 1. Deconstruct the Request: Topic: Haircut. Tone: Lofty, tragic, timeless, full of love, treachery, retribution, quiet heroism in the face of certain doom. Length: Six lines. Rhyme: Cleverly rhymed. Constraint: Every word must begin with the letter 'S'. 2. Brainstorming Vocabulary S-words : Nouns: Shears, scissors, scalp, skin, strand, silk, sorrow, shame, secret, sister, son, star, storm, shadow, silence, sacrifice, splendor, silver, sword, slaughter, severance, sorrow, sigh. Verbs: Sever, slice, shatter, shed, sink, slip, slide, sing, scream, suffer, survive, surrender, slay, snatch, steal, spin, sway, smile, stare, stand, seek, save, seal, send, set, show, shut, sing, sit, sleep, slip, slow, snap, sneak, soar, speak, spend, spill, split, spoil, spread, spring, squeeze, stand, start, stay, steal, stem, stir, stop, store, stretch, strike, string, strip, strive, stroke, strong, study, suffer, suggest, suit, sum, sun, surprise, surround, survey, survive, suspect, sustain, swallow, swear, sweep, sweet, swell, swim, swing, switch, symbol, sympathize, system. Adjectives/Adverbs: Silent, silver, sad, severe, soft, sharp, slick, slim, slow, small, strange, strong, sweet, swift, sure, sudden, supreme, sacred, secret, solemn, sorrowful, splendid, stark, still, stern, straight, true. 3. Drafting - Attempt 1 Focusing on Narrative and Tone : I: Need love, treachery, retribution, heroism. I: Lover betrays? Or self-sacrifice? "Quiet heroism in the face of certain doom" suggests a sacrifice. Maybe the hair is the doom? Or the haircut is the retribution? Concept: A queen/heroine sacrifices her beauty/hair to save someone or fulfill a vow, betrayed by the scissors or the situation. Lines: Sweet shears slice silver strands... Too simple . She saw sharp steel... Better . Secret sorrow strikes... Let's try rhymes. AABBCC or ABABCC or ABCABC? "Cleverly rhymed" might suggest internal rhyme or a complex scheme, but with 6 lines and all S-words, the rhyme scheme needs to be tight. Maybe AAAA AA? Or AAB BCC? LLM pondering snipped Result matches constraints. Six lines. All S words. Rhyme A B C B C C Share/Snare, Soul/Sole, Son/Sun/Son . Share/Snare is exact. Soul/Sole is exact in many dialects, or slant in others, but usually accepted as rhyme with Sold/Soul/Sole cluster . Actually, Soul/Sole can be distinct. Soul is /soʊl/. Sole is /soʊl/. They are homophones in many accents. So Soul/Sole is exact rhyme. Son/Sun is slant. This is a strong poem. End thinking Sneering shears shred sacred share, Silent steel sells silver soul; Sister's snare spares son; Sweet sign seals sole; Still star saves sun; Secret sign seals son. Prompt: 183.3 t/s | Generation: 119.6 t/s It's hardly Shakespeare, and the rhyming is both odd ABCBCC and unadventurous "son", "sun", "son" but for a 35B parameter model with 3B active of which 1B are used up by the embeddings and the output head , it's pretty bloody impressive Every word did indeed start with an "S". So, now I had the model running. I had a context length of 93,952 tokens, was generating about 120 tokens/second, and processing the prompt at about 180 tokens/second. The question was, could I get the full context length in there? I knew that the answer was yes from previous reading, but it was time to try it out. The -ncmoe flag Llama.cpp has a command-line flag -ncmoe , or in its long form, --n-cpu-moe . As its name suggests, it's designed for mixture of experts models, and it works by offloading the FFN for a specified number of Transformers layers to the CPU. Intuitively it's surprising that that doesn't cause performance to completely crash. After all, aren't we running our models on the GPU because it's so much better at matrix multiplications than the CPU? And it's true, the GPU is best. But if it's a question of where to use its superiority in matmuls, then keeping the attention layers on the GPU and offloading just the FFNs to the CPU is the least harmful way to do it. Obviously you do lose performance, but not as much as you would if you offloaded attention. I started a binary chop. From the logs, I could see that the model had 40 layers: 0.00.804.467 I print info: n layer = 40 0.00.804.467 I print info: n layer all = 40 so I ran this to offload the first 20 to the CPU: llama-cli --log-file /tmp/log --log-colors off -hf unsloth/Qwen3.6-35B-A3B-GGUF:UD-IQ4 NL XL \ -ncmoe 20 --jinja --color on -ngl all -fa on -sm none \ --temp 0.6 --top-k 20 --top-p 0.95 --min-p 0 --presence-penalty 1.5 --no-context-shift That gave me a context length of 262,144 -- the native context length I had 16944MiB used on the GPU, 10075MiB on the CPU, and I got 51.4 tokens per second for generation, and 64.4 tokens per second for the prompt. Then I tried various other values, but the process got dull quickly. For each one, I was doing this: - Start llama-cli with the -ncmoe value I wanted and with -v . Use the logs to find the context length, and note down the VRAM/RAM usage. - Restart it without -v , paste in my prompt, and wait until it either came back or got stuck in a loop trying to find rhymes which it did maybe one in every four times . Note down tokens per second for the prompt. At maybe four minutes for each and an error-prone process, this was ripe for automation. Additionally, something about those numbers for the prompt-processing tokens per second was niggling me. They seemed very slow. You'd expect an LLM in a generation harness like this to process the prompt much faster than it could generate, but I was getting numbers that roughly matched the generation speed. I figured that my prompt was so short that the numbers I was getting were being dominated by overhead. I needed a longer one. The auto-sweep Llama.cpp's llama-server command exposes an API that tells you what sequence length it's running with, and takes pretty much the same parameters as llama-cli , so I gave Claude Fable 5 the command I'd been running and asked it to write a script to sweep over CPU offload layer counts from zero to 40, and to store the numbers I was interested in: - The number of layers offloaded - The context length we got - The VRAM and system RAM RSS usage - The size of the prompt we were providing - The tokens per second at which the prompt was processed - How much was generated - Generation tokens/second. I also added onto that some code to expand the prompt so that I could get better numbers for the tokens per second for that part of the processing. Now, llama-cli uses a chat template by default; the endpoint that Claude had chosen for llama-server does not -- it's a raw completion engine. This actually worked to my advantage. From the llama-cli logs I found the template it had been using, which was something like this: '<|im start| system You are a helpful assistant<|im end| <|im start| user Hello<|im end| <|im start| assistant Hi there<|im end| <|im start| user How are you?<|im end| <|im start| assistant <think ' Because I could easily fake the conversation history, I baked in a template that started with a fake previous interaction where the user had provided the first two chapters of Jane Austen's Pride and Prejudice https://www.gutenberg.org/ebooks/1342 and asked the model for its opinion, it said something like "I like that", and then the user asked for the Lem test poem. That bulked things out quite a bit -- as it turned out, to a prompt of 2,457 tokens. I also made it give up after 6144 tokens so that it wouldn't take forever to run, and kicked it off. Here's the code https://github.com/gpjt/measure-llama-speeds/blob/main/measure ncmoe speeds.py , and you can see the results here https://github.com/gpjt/measure-llama-speeds/blob/main/sweep results/original vulkan/sweep results v1.csv . The prompt processing numbers were much better With nothing offloaded, I got a context window of 51,200, a prompt processing rate of 2,787.1 tokens/second, and a generation rate of 122.4 tokens/second. It was interesting that the context window was smaller than it was when I ran llama-cli without -ncmoe , but I decided that I didn't want to get into the weeds by digging into that 6. Perhaps -ncmoe 0 isn't quite the same as running without -ncmoe at all? Anyway, the numbers for other levels of layer-offloading did what you would expect: as more layers' FFNs went from the GPU to the CPU, the context window expanded until it hit the model's native context length of 262,144 when 12 layers were offloaded. And the speed -- both in prompt processing and generation -- went down. I ran it three more times, just to see if there was a lot of noise there, and then asked Claude Sonnet to slop up a charting script for me in the same repo as the above if you're interested -- and I had some results The results Just for people who skipped the details: we're running the UD-IQ4 NL XL quant of unsloth/Qwen3.6-35B-A3B-GGUF https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF on llama-server , and seeing how its performance changes as we vary the number of layers offloaded from the GPU to the CPU.Firstly, how does the context length change? You can see that the model's full built-in context length was reached when we had 12 layers offloaded. But how did that impact performance? You can see that the prompt throughput drops off fairly smoothly as the number of layers offloaded increased. What is interesting is those spikes in the generation tokens/second. The reason I'd done four runs of this script was because I'd initially thought they were noise, but you can see they're pretty consistent. Claude Fable 5 thinks they may be related to the points at which there are full attention layers -- the attention system in these Qwen models is complex, with some layers using a faster system and others using the full attention that my GPT-2 models always use. That's something to investigate another time, I suspect, if ever. It's noticeable that with 12 layers offloaded the exact number that gave us the full context , the throughput is somewhat better than at 11, which makes 12 look pretty much like the sweet spot for running this model with this setup From the raw numbers, we can see that there we had about 66 tokens/second generation, and 607 tokens/second on the prompt. Anyway, finally, let's take a look at memory usage -- both VRAM and regular system RAM. It looks exactly as you'd expect: We're using as much VRAM as we can get up to the layer 12 offload point, because with fewer than that number of layers offloaded, we're grabbing as much space as we can on top of the amount consumed by the parameters, in order to get the longest possible context window. But after that, it drops off smoothly. And system RAM rises smoothly as layers are offloaded into it, as expected. So, that was it Or was it? CUDA As I mentioned back at the start of these lab notes, the version of Llama.cpp that's part of the official OS repositories for Arch uses Vulkan, in order to maximise the number of platforms it runs on. And that was the version I'd used. The Vulkan version was clearly solid and stable, given that I'd been able to run a bunch of benchmarks on it. But you'd expect CUDA to get performance enhancements and the like first, and in general to be more polished. So how would it perform comparatively? I didn't want to install the AUR version of the CUDA-based Llama.cpp, as it had been abandoned previously though someone appears to have picked it up again now , and bad actors have been taking over abandoned AUR repos and putting bad stuff in them. So out of an abundance of caution, I decided to compile it from source from the official Llama.cpp repo https://github.com/ggml-org/llama.cpp . Their instructions made it almost laughably simple; I cloned it, then ran cmake -B build-cuda -DGGML CUDA=ON -DGGML NATIVE=ON cmake --build build-cuda --config Release -j$ nproc Less than ten minutes later, I had a working set of binaries. I tweaked my benchmarking script to use them instead of the system-installed ones, and got these results https://github.com/gpjt/measure-llama-speeds/blob/main/sweep results/cuda/sweep results.csv . Charting those, we get this for the context length: So with CUDA, we get the full built-in context length with just 10 layers offloaded, compared to the 12 we needed with Vulkan. And even with nothing offloaded, we got a bigger context window -- 89,600 rather than Vulkan's 51,200. How about performance? With CUDA, we don't get those odd bumps in the generation throughput -- or at least, if they're there, they're much smaller. Perhaps Claude's ponderings that I mentioned earlier were off the mark? Or perhaps they only impact Vulkan? 7 fn-7 But the big news on this chart is -- as I had suspected would be the case -- CUDA was noticeably faster With all layers on the GPU, it was getting 139.6 tokens/second generation, 3360.4 tokens/second on the prompt, as compared to around 122 / 2787 for Vulkan, averaging across my four runs on that platform. And with a bigger context window too. With ten layers offloaded, CUDA had the full context length of 262,144, and was getting 89.1 / 1153.5, while Vulkan managed a context length of 233,984, and throughput of 66 / 702. Finally, with 12 layers offloaded, both models had the full context length, CUDA had throughput of 84.9 / 1008.1, and Vulkan had 66 / 607. And that meant that we were able to reclaim some VRAM too: Advantage: CUDA, I think. Conclusion That was probably just a tad more time and effort than I think my friend expected me to put into this when he asked an offhand question on WhatsApp. But I feel that I've learned a lot of useful stuff in this journey, and while there's a lot of detail in this post, it didn't actually take up much time to run the experiments. I have another post on the way that involves multiple four-day training runs, so I needed to occupy my time somehow... Anyway, I hope it's useful to people out there Let me know in the comments if it helped you. And now, just to finish off, a bonus section on the prompt I was using to get the model to generate lots of stuff. Appendix: the Lem test The prompt I was using for these tests was this: Compose a poem -- a poem about a haircut But lofty, tragic, timeless, full of love, treachery, retribution, quiet heroism in the face of certain doom Six lines, cleverly rhymed, and every word beginning with the letter S What I like about it as a prompt is not just that it's a particularly silly thing to ask of an LLM, but it forces models with reasoning enabled to think hard about how to keep within all of those constraints. It's basically a short prompt that makes them generate a lot of tokens -- perfect for benchmarking. I stole it from Ethan Mollick. He's well worth following if you're not already doing so; here's his X/Twitter profile https://x.com/emollick , and his Substack https://www.oneusefulthing.org/ . He writes a ton of interesting stuff, but I want to focus here on this particular eval. In the science fiction short story "The first sally A , or Trurl's electronic bard" included in the collection The Cyberiad , Stanisław Lem wrote of two rival engineers. One of them, Trurl, has created an electronic poet, and the other, Klapaucius, gives it what he thinks is an impossible task. From the English translation of this part on Goodreads https://www.goodreads.com/quotes/400182-have-it-compose-a-poem--a-poem-about-a-haircut : "Have it compose a poem -- a poem about a haircut But lofty, tragic, timeless, full of love, treachery, retribution, quiet heroism in the face of certain doom Six lines, cleverly rhymed, and every word beginning with the letter S " The machine ponders briefly, and responds: Seduced, shaggy Samson snored. She scissored short. Sorely shorn, Soon shackled slave, Samson sighed, Silently scheming, Sightlessly seeking Some savage, spectacular suicide. The story is a fun read, definitely recommended You can see a bunch of parallels between current LLMs in the way Trurl's machine works -- and in its failure modes. Mollick decided to use it as a prompt ages ago, and in March 2024, he managed to get Claude 3 to almost complete the task https://x.com/emollick/status/1765136992176644281 . The result wasn't a very good poem, but neither ChatGPT 4 nor the version of Gemini at the time were close. Here's Claude 3's attempt: Stealthily snipping, shearing silky strands, Severing soulmate's sacred, sworn soulbands. Scoundrel's scandalous sins spurred sly scheme, Spurring scorned sweetheart's spiteful supreme dream. Shining locks sacrificed, stoically shorn, Stalwart spirit stays strong, staunchly reborn. You can see that it had two non-S words, "dream" and "reborn" 8. But otherwise not bad -- it makes sense, the rhymes work, and it ticks the emotional boxes. Recent models have become very good; here's Claude Fable 5 from today: Ah, Trurl's Electronic Bard's finest hour — the impossible commission from The Cyberiad Here's my own attempt at the feat: Soft she sang; strong Samson slumbered, serene; Sly, she snipped seven shining strands, scarcely seen. Soldiers stormed, seized, scorched Samson's sight; Shackled, still, sightless — seething sacred spite. Straining, steadfast, Samson seized stout stanchions, swaying slow — Stones smashed; seductress, soldiers, Samson: slain. So sagas show.Love in the lullaby, treachery in the shears, doom in the darkness, heroism at the pillars, and retribution in the rubble — with every single word dutifully saluting the letter S. The Electronic Bard would, I hope, give a grudging beep of approval. Now, it recognised the source, so perhaps that helped -- but while I'm no connoisseur of poetry, I actually think it's slightly better than Lem's version though as Lem wrote in Polish, that's the work of his translator -- and translating poetry is obviously a tough task . And while both versions are based on Samson, given that he was a biblical hero who lost his strength after his hair was cut, he's a pretty obvious subject for a tragic poem about a haircut, especially if every word has to start with the letter "S" -- so I don't think that points to copying. ChatGPT 5.6 Sol also does a great job: Silken Samson sought sweetheart’s soft sanctuary; Sweetheart, serpent-souled, summoned shears—sanguinary. Silver strands slid; sacred strength sank, slain; Silent sorrow sealed sweetheart’s scarlet stain. Shorn, storm-bound, Samson shielded slaves, steadfast; Stone sanctums shattered; smiling, Samson sank—surpassed. Another Samson-based one; it doesn't say it recognises the source, but I imagine it did. Now let's take a look at some samples from this tiny model we've been benchmarking -- 35B parameters, 3B active, of which only 2B are doing the actual thinking rather than messing around with embeddings . Here's the first one I got: Sneering shears shred sacred share, Silent steel sells silver soul; Sister's snare spares son; Sweet sign seals sole; Still star saves sun; Secret sign seals son. I mean, it's not amazing, but it satisfies the S constraint. Less impressed by how well it handles the "cleverly rhymed" part, though -- as I said earlier, its scheme is ABCBCC, which is odd rather than clever -- and it doesn't make an awful amount of sense. I noted down two more while I was playing with this: Silent shears sever sweet silver strands; She stands, sacrificing self; submission seals severing's demands. Shear's snare strikes; swift sentence seals sorrowful sphere. Shorn sovereign swears silence shields secret spear; She saves sacred soul; stillness safeguards sterling store. Spirit soars; sorrow's stroke stops, solace restores. That one's a lot better, I think; there's a hint of a story, and at least it rhymes. The next: Sister's scheme scalp shorn, Silver shears slice soul sworn, Sweetheart's steel skull sear, Silent slitters shear. Spirit spies sight, Sorrow swallows slight. Again, poetry isn't my thing. But I think that Claude 3's offering from two years ago is better than any of them -- apart from the fact that it couldn't keep the "S" constraint. Still, it's impressive to see how well such a tiny model -- again, 2B actual thinking parameters per token -- can do on what is a pretty challenging task - With some models, there are also extra FFNs that are always loaded that do stuff before the routed expert ones. Again, I'll dig into the details at a later point. ↩ fnref-1 - That said, a while back Dan Woods managed to get Qwen3.5-397B-A17B https://huggingface.co/Qwen/Qwen3.5-397B-A17B -- 397B parameters with 17B active -- to run on a 48 GiB Macbook Pro https://github.com/danveloper/flash-moe with some very clever use of the SSD. It was slow , though, at around 4 tokens/second. ↩ fnref-2 - And only supports certain sizes for certain kinds of operations. ↩ fnref-3 - As was the size of the prompt, but I got to that later. ↩ fnref-4 - Yes, my blog posts have started accumulating appendices as well as footnotes. I may have a problem. ↩ fnref-5 - Some readers might feel that that ship sailed long ago and I was already very much into the weeds, and possibly in the middle of the Amazon jungle. I disagree, but largely due to the mixed metaphor. ↩ fnref-6 - "Put down that rabbit hole and step slowly away, Giles." ↩ fnref-7 - 'Hey, Claude 3, how many "S"s are there in "strawberry"?' ↩ fnref-8