• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Recent content by bwang

  1. B

    Anyone a gambler? Konica Minolta 2300dl new in box. Made circa 2004 (that's when I got mine)

    I am so confused. Why is this worth $600 shipped?
  2. B

    What LLM/Model you using?

    arena is A/B comparisons in short form chat which tends to favor things such as style, brevity, and formatting. it also doesn't test multi-turn tool call performance, error recovery, or planning which are important for coding.
  3. B

    Hermes - Your new best friend

    oof, come on Nous who is doing your marketing, I literally cannot tell what the new feature is from this post. It seems like they've added the ability for the main session to chat with the sub-sessions? That's a good feature - it prevents main session context from getting poisoned with failed...
  4. B

    Guide to running big models on a big hobbyist budget

    There is something reassuring about being able to run frontier models at home even if it makes no financial sense. To achieve this, we describe two configurations capable of serving GLM-5.2/5.3 in a modest (sub-$20K) budget that fit in 220V 50A residential service. The goal was to achieve...
  5. B

    Arc Pro B50 Guide and Discussion feat. bonus! B50 without resizable BAR

    What is it? The Arc Pro B50 is a unique bus-powered low profile card with 16 GB of VRAM. It is very slow: 85 TFLOPS bf16 XMX throughput, 170 TOPS int8 XMX throughput 224 GB / sec of memory bandwidth However, "slow" is relative - the B50 has more throughput and bandwidth than a $5000...
  6. B

    This Sub Going Forward - Your Suggestions?

    Doing things a bit differently: - Client hardware + models: stuff which runs on your existing gaming rig or something that looks like one. Good idle, OK cost, and the computer still usable for other things. All the Qwen-27B people will wind up here - Datacenter hardware + models: stuff which...
  7. B

    Cooling an AMD Instinct MI210 for local inference

    if it is anything like the nvidia stuff figure max power during prefill for a couple minutes at a time, 80w during decode, and max power generating images/videos
  8. B

    Cooling an AMD Instinct MI210 for local inference

    the static pressure drop across the server cards is insane. have Claude code you up something that does full-RAM-sized bf16 GEMMs or spin up ComfyUI and have it generate some long videos with Minimax-H3 and keep an eye on the HBM temps as you tune cooling, that is a $4000 card so I'd err on the...
  9. B

    AI LLM hardware discussion thread

    I just spent all Saturday chasing a 1-in-a-million ghost serving GLM-5.2 on my new 640 GB system (10x bootleg A100) and let me tell you, I've never hated computers this much... As for actual system specs, see sig.
  10. B

    Wtb: 3080 3090/ fe / ti, 6950X i7 cpu, & DDR4 3600+ sets 4x4gb min

    what's the goal here? collectibles or building a working rig? it sounds like you want the original packaging out of a Rampage or Extreme11; that's gonna be really hard to come by ten years in lol. arguably the box is worth more than the motherboard itself now, because I can't imagine anyone...
  11. B

    FS: GSkill DDR 5 Memory

    ygpm
  12. B

    FS HP Zbook

    ygpm on the laptop
  13. B

    Can RTX-4090's still be found brand new?

    Because 4090's are made on the same process node as 5090's. If they kept making 4090's, they'd be throwing away $500 for every one they made. Also, let me know where you're finding $2200 5090's because I'd like to buy a few :)
  14. B

    AI Markets Were Deceived To Believe In DeepSeek’s Low Training Costs; They Are Actually 400 Times Higher Than The Reported Figure

    This is hilarious Nvidia shilling. DeepSeek's capex into the hardware they used for training was a lot of money, but the actual GPU hours they burned in training this specific model was only a few million dollars, as calculated against current cloud pricing (it's actually less if you only count...
  15. B

    RTX 5090 Actual Performance Improvement - 6 FPS in CyberPunk

    Blackwell's success will be predicated on whether neural rendering actually works for gaming. I can speak from experience as an engineer in the field that the general technique is real and should be taken seriously - basically, the idea is rather than waiting for frames 5 and 6 and then using...
See more
Back
Top