7 ms·
DeepSeek V4 Flash on a Single AMD MI300X
- majke 2mo agoI don't think you can buy a single "MI300X" unit, right? Only the box with x8 of these at a cost of ~250K EUR.
- baalimago 2mo agoGive it an AI-bubble pop and these will be flooding the market.
- _factor 2mo agoThey will be instantly bought out by companies, not individuals. The consumer bubble won’t pop for quite a while yet. Production also won’t ramp up while lack of real competition keeps the demand high.
- aurareturn 1mo agoWhen is it popping? Is the AI bubble in the room with us now?
- baalimago 1mo agoNext month perpetually
- amrit3128 1mo agoTomorrow? Next year? In 5 years? Nobody can say. But we do know that AI is overvalued, so it WILL pop.
- aurareturn 1mo agoWell, if no body can say when it will pop, then can we really say it's a bubble and it's overvalued? I can tell you it will pop in 10 years and when it pops, it will still be 20x bigger than in 2026. Does that even make any sense? People said AI bubble will pop soon in 2024 and that it was overvalued. Turns out, many AI stocks 10x, 20x since 2024. Actual usage has gone exponential as well. Anthropic revenue went from $100m ARR at start of 2024 to $80b ARR today.
- Joel_Mckay 1mo agoMany are saying July 2027, as in the past these market corrections have correlated with Shrek movie releases. Debt-backed investors have to pay up eventually. =3
- ekidd 1mo ago> Well, if no body can say when it will pop, then can we really say it's a bubble and it's overvalued? Well, given the literal trillions being spent, the only ways this pays off are: 1. AI replaces a non-trivial fraction of human employees. 2. Someone builds a Culture Mind, and humans become (hopefully) pampered pets of AIs we don't understand. Seems unlikely, but it would arguably count as a payoff even if it made money meaningless. Or maybe the AIs don't want pets, and you get SkyNet. Which definitely doesn't care about paying off anyone's investments. When you look at various news articles about investors, yeah, there are definitely a lot of rich people who think that they're going to automate all human labor or just bring about the Singularity. Possibly with them in charge of the rest of us. If you don't make these kinds of wild assumptions, then yeah, this is looking like one of the biggest bubbles ever.
- aurareturn 1mo agoCan we see some actual numbers, projections, models instead of vibes?
- atwrk 1mo agoDebt is at $3 trillion right now: https://fortune.com/2026/07/31/ai-debt-hypescalers-capex-capital-spending-hidden-borrowing-bond-issuance/ https://fortune.com/2026/07/31/ai-debt-hypescalers-capex-cap... Interest alone, at assumed 5%, amounts to about $150 billions per year. That's probably higher than the combined AI revenue of the top 3 providers.
- aurareturn 1mo agoDid you read your own article? It's well less than $3 trillion. Now let's build the model out more. What is the projected revenue, backlog, improvements in existing big tech businesses such as AI helping Meta's ad business?
- slaw 1mo agoThe AI bubble will pop when China gets access to EUV, so the earliest it could happen is 2030
- dghlsakjg 1mo agoNvidia has ever so slightly underperformed the SP500 YTD (at the exact time this comment is being typed), so its basically the apocalypse already.
- tamimio 1mo agoThing is, GPUs will always be on demand, look at their history, initially for gaming, then for hash cracking, then 3D rendering, then for crypto mining, and now AI training and fine tuning. When AI bubble bursts, there will be another bubble taking over. The only solution is more companies making high end units, only competition will make it better for consumers.
- segmondy 1mo agono they won't , the bubble is a financial thing. the demand is real and not going away.
- atwrk 1mo agoThe big question is whether the demand will stay if the subsidized pricing ends. That's what the bubble talk is about. Right now all the players compete for market share and don't care about the losses (hence the debt). But what happens if no one wants to lend them anymore?
- jack_pp 1mo agoI don't think inference is subsidized, it's the training. So what happens is, there's no new models anymore or are released slower.
- FeepingCreature 1mo agoAPI inference is probably not subsidized. Coding plans absolutely are.
- vehemenz 1mo agoGood distinction. The demand is partially driven by the low costs, which are only low because the major providers are losing money.
- qwytw 1mo agoThere is no evidence they are losing money on inference, though? Also if they are keeping the price low because they want to gain market share and reduce the competitiveness of Chinese models they won't be able to raise prices without providers serving open models (at cost + low margin) severely undercutting them.
- vehemenz 1mo agoThere's no evidence they're making money, and we already know from the projected datacenter capacity in a few years that there will be for more supply than demand, so the major providers will have to repay that debt. Even if they are making money on inference, it's nowhere near enough to cover the bill. It's a losing proposition either way, especially with Chinese models now in play.
- zhoutong 2mo agoIt’s available on demand from a few cloud providers. Seems like the cheapest is AMD Developer Cloud (https://www.amd.com/en/developer/resources/cloud-access/amd-developer-cloud.html https://www.amd.com/en/developer/resources/cloud-access/amd-...) powered by Digital Ocean at $1.99/hour. Edit: Now I think about it, this might be the cheapest way to run the DeepSeek V4 Flash 0731 on a dedicated inference server at original weights. I haven’t run mixed load benchmarks but I guess it’s possible to generate $3-$4 worth of tokens per hour and still maintain a usable per-user throughput.
- WASDx 2mo agoAt 830tok/s * 1 hour that's almost 3M tokens which is just $0.54 worth of tokens at Deepseeks current output price.
- krisknez 1mo agoHow is that economically viable? They are selling at a loss?
- simlevesque 1mo agoThey get all our invaluable data which they'll use to train the next model, to get more data, to train the model after.
- jingpostmedia 1mo ago[flagged]
- dietr1ch 1mo agoThey claim their advantage is knowing how to serve their models efficiently, which is quite possible since they design for it.
- gpugreg 1mo agoAgentic workloads are somewhere around 1%/0.5%/98.5% input/output/cached tokens. Cached tokens are pretty much free for inference providers (if they implement sparse and compressed attention properly) and throughput for input tokens is much higher. Lets assume that you've got 2 million input tokens, 1 million output tokens and 98.5 million cached tokens to process. That would cost 2 * $0.14 + 1 * $0.28 + 98.5 * $0.0028 = $0.8358 with DeepSeek API pricing. For comparison, it would take 2M / 8000 + 1M / 800 = 1500 seconds to process this amount of tokens with the linked framework, which is about $0.83 when we assume $2/hr for one MI300X. However, other inference providers have 10 times higher prices for cached tokens, which results in a comfortable margin. And we should not discount that DeepSeek also gets paid in data, which is probably more valuable to them. And I believe that this framework still has some room for optimization for generation with high batch sizes.
- Lwerewolf 2mo agoThe MI350p exists and should run a decent quant (say, the ~96GB antirez mix) well, but you can get two rtx pro 6000s for one of these, or 8x (actually more) r9700 + probably the gear to run them, etc. Otherwise, you can probably buy one of these second hand from somewhere (SXM A100s are available that way) and run it in an adapter board.
- touisteur 1mo agoI thought MI350P wasn't available yet, curious where to source it right now.
- FuriouslyAdrift 1mo agoThere's at least one systems integrator selling a rack server with 2x MI350Ps. I haven't seen the cards all by themselves yet.
- _joel 1mo agoI thought it was a consumer grade GPU until I saw the 192GB of HBM and 256GB or RAM.
- varispeed 1mo agoTo be fair the development of GPUs have stalled over the years. If they kept up with the progress instead of focusing on enterprise market, likely 256GB consumer GPU would be a norm today.
- FuriouslyAdrift 1mo agoIt's a chopped down MI350X (roughly half the performance)
- FuriouslyAdrift 1mo agoBasically right from Lisa Su's speech: "AMD is essentially taking one of its MI350X accelerators and cutting it in half, resulting in a card with half as many compute resources, half as much memory, and perhaps most importantly, a bit over half of the power consumption"
- throwawayffffas 1mo agoYou can get one on ebay for like 20k, but it comes without the backplane and i dont think there is a pcie card adaptor from china like the ones for h200.
- jkwang 2mo ago[flagged]
- xorfish 1mo agoThis is still quite a bit away from the performance that deepseek gets on their H800. In their DSpark paper they report a throughput of 15k tokens/s/gpu. The MI300 should be able to compete with the H800 so there are probably still quite a few optimizations that can be made.
- somnial 1mo agothroughput scales superlinearly with number of GPUs when networked well and deployed with wideEP, so 1x won't compare. also it would be interesting to figure from the DSpark paper whether their numbers are consistent with the GPUs still being H800s, since they never actually say...
- hn0tdqaek4 1mo ago[dead]
- Tepix 1mo agoUnfortunately, the MI300X is an OAM module. The MI350P is the one you want: It's a PCIe card, but it has less memory: 144GB. Luckily, DeepSeek V4 Flash will run in 144GB too because it's 256 MoE exports are native MXFP4 quantized.
- WhitneyLand 1mo agoHow do you figure that? When they just loaded the weights alone, it was taking 156GB in vLLM. After warm-up and adding a KV cache pool, it took over 200GB. And this implementation is already cutting down the 1M token context window you would normally get.
- Tepix 1mo agoFor sure if you want to properly utilize the model with several users in parallel and large context you'll want two MI350P.
- craftkiller 1mo agoJust want to add that while the MI350P is a PCIe card, it is designed for servers. It just has a heatsink (with no fan) which the powerful full-case fans of a rackmount server are supposed to cool. So while the MI350P is certainly more attainable for us regular folk due to its formfactor, we won't be able to just drop it into our gaming PCs like a regular graphics card. That being said, if you're dropping tens of thousands of dollars on graphics cards then picking up a rackmount case to go around the card is pretty insignificant.
- cyberax 1mo agoJust put a fan on it. It's just 600W, so nothing super-special is needed. Or add a water cooler.
- monster_truck 1mo agoYou're going to need at least a couple loud ass IPPC 3000s to usefully move that kind of heat if you don't want it to throttle. And then another normal sized fan for the doorway of the room it's in. Not exactly super special, but a ~constant 600W+ of heat tends to be a learning experience. It's worse than a high end gaming rig, much closer to a literal space heater. I don't work during the summer because it sucks fighting both this and the sun with AC. I do have a fan that slots into the window and can push or pull, but kicking the waste heat outside doesn't help when its humid.
- sylware 1mo agoIs their hardware programming interface reasonable for implementing inference of frontier models: no quantization, several tera params? BTW, how many many params open weight frontier models have? A few teras, 100s of teras?
- wren6991 1mo agoKimi-K3: 2.8T Qwen3.8-Max: 2.4T DeepSeek V4 Pro: 1.6T DeepSeek V4 Flash: 284B (all are total parameter counts, not active parameters)
- sylware 1mo agoRumors say chatgpt/claude/gemini/etc are in the 100s of teras. True?
- wren6991 1mo agoI'll ask my uncle (he works for Nintendo) and get back to you on that one
- sylware 1mo agoMy question is that wrong?
- wren6991 1mo agoSorry if the joke didn't land; I have heard a lot of different numbers for the size of US labs' models, but never seen any of them substantiated, so I think you're likely to just get more rumours in answer to this question. My personal take, with no sources: 100T sounds excessively high given they need to be able to actually serve these things on commercially available hardware. I would guess they are in the same order of magnitude as the Chinese frontier models. It's possible their edge is in RL training methods, training-time compute, and access to data (e.g. from customers' CC/Codex sessions), not in model size.
- GTP 1mo agoStrange that in the prior art they didn't list DwarfStar, as it is able to run the same model (probably quantized differently though) in less memory. Maybe the author isn't aware of it?
- WhitneyLand 1mo agoAnother headline of “model runs on x”, which usually means “let’s list how much you give up to run on x”. Dumbed down quantization? No. Full intended inference weights preserved, so far so good. Slow performance? No again. Looks like you could get over 150 tokens/second. Give up context window size? Yes. Original model is trained for and served at 1M, this is 256k. A very practical tradeoff though. Codex is in this range, and quality does start to drop off toward the full size.
- bwfan123 1mo agoI am curious if there has been work to remove experts from an open-weights model. The goal would be to reduce the size to be able to run on desktop GPUs without compromising quality. For a focused usecase - say coding, you dont need a model that knows world history. And, I am not talking about quantization. If it is possible to determine which experts are active for some usecases, and surgically remove the others.
- monster_truck 1mo agoGlazing over a lot, that's how they work already, just not in the way you think. A relatively small fraction of the model is active at any given time
- smallerize 1mo agoExperts aren't trained on separate tasks. More recent routers are designed to spread out requests even more evenly, and they were already pretty even.
- Tepix 1mo agoYes. It's called REAP and from what I've seen, results aren't stellar.
- monster_truck 1mo agoIn my experience the 1M context is genuinely too much. The first time I swapped from OAI to DSv4P, I checked and double checked that the harness/etc was working correctly over the course of hours and hours of work thinking that I had set something up wrong because it simply never had to compact! The drop in quality is arguably less than that of what you get from compact to compact on Codex, which is good for what it is or was. Was also surprised to learn just how much of Codex's window was being burnt on shit I didn't want or use. Sure I can pass this and that flag to eliminate most of it, but for a $200/mo product aimed at professionals, that isn't something anyone should have to janitor (also totally ignoring the bandaid of banked resets they've slapped over their repeated mistakes). It's wild just how far $20 will get you with Deepseek, even at their new rates. Buyers Remorse is my very least favorite feeling, I felt sick thinking about what the $1200 I had given OAI this year would have gotten me had I only tried sooner.
- fergusfinn 1mo agonice! i think the higher HBM on Mi300x is really useful for this kind of thing we did some work on this for 2xMi300x (kindly referenced in the readme) https://blog.doubleword.ai/deepseek-v4-flash-mi300x https://blog.doubleword.ai/deepseek-v4-flash-mi300x. https://hotaisle.xyz/quick-start https://hotaisle.xyz/quick-start hotaisle is great for getting Mi300x to experiment with
- pop3zxcv 1mo ago[dead]
- PrimeAli 1mo agoGreat