DeepSeek V4 Flash on a Single AMD MI300X

(github.com)

81 points | by zhoutong 2 hours ago

2 comments

  • majke 1 hour ago
    I don't think you can buy a single "MI300X" unit, right? Only the box with x8 of these at a cost of ~250K EUR.
    • zhoutong 1 hour ago
      It’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-...) 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 57 minutes ago
        At 830tok/s * 1 hour that's almost 3M tokens which is just $0.54 worth of tokens at Deepseeks current output price.
        • lnenad 10 minutes ago
          830t/s is burst aggregate. ~500 is sustained and it's for 8 concurrent users. Meaning for $1.99/hour if you serve 8 users it's 8*$0.54, not just $0.54.

          You shouldn't rent one out if you're just serving it for yourself, but from a financial standpoint if you sell to users you can take a 100% margin.

        • Almondsetat 4 minutes ago
          You get privacy for 4 times the cost
        • krisknez 20 minutes ago
          How is that economically viable? They are selling at a loss?
          • pama 1 minute ago
            Use nvidia hardware instead and use a larger cluster serving many more users concurrently. Easily 10x–20x higher token rate per GPU with public solutions like dynamo and sglang.
          • gpugreg 5 minutes ago
            Agentic 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.

            Also, I believe that this framework still has some room for optimization for generation with high batch sizes.

          • dietr1ch 14 minutes ago
            They claim their advantage is knowing how to serve their models efficiently, which is quite possible since they design for it.
        • thrownaway561 12 minutes ago
          This is exactly what I came to say. The price of Flash is so cheap that trying to run it locally or with your own hardware is pointless. I was using it about a month ago to program some stuff and ran it for 4 days non-stop and it cost me about $2.
      • langs 41 minutes ago
        You need to optimize the KVCache part(save to disk to save compute) to achieve this goal.
    • _joel 4 minutes ago
      I thought it was a consumer grade GPU until I saw the 192GB of HBM and 256GB or RAM.
    • Lwerewolf 43 minutes ago
      The 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.

    • baalimago 1 hour ago
      Give it an AI-bubble pop and these will be flooding the market.
      • _factor 1 hour ago
        They 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 34 minutes ago
        When is it popping? Is the AI bubble in the room with us now?
        • baalimago 11 minutes ago
          Next month perpetually
  • jkwang 1 hour ago
    [flagged]