top | item 44791860 (no title) rapfaria | 7 months ago Not sure what you mean or new to llms, but two RTX 3090 will work for this, and even lower-end cards will (RTX3060) once it's GGUF'd discuss order hn newest axoltl|7 months ago This isn't a transformer, it's a diffusion model. You can't split diffusion models across compute nodes. karolist|7 months ago do you mean https://github.com/pollockjj/ComfyUI-MultiGPU? One GPU would do the computation, but others could pool in for VRAM expansion, right? (I've not used this node) AuryGlenz|6 months ago Nah, that won’t gain you much (if anything?) over just doing the layer swaps on RAM. You can put the text encoder on the second card but you can also just put it in your RAM without much for negatives.
axoltl|7 months ago This isn't a transformer, it's a diffusion model. You can't split diffusion models across compute nodes.
karolist|7 months ago do you mean https://github.com/pollockjj/ComfyUI-MultiGPU? One GPU would do the computation, but others could pool in for VRAM expansion, right? (I've not used this node) AuryGlenz|6 months ago Nah, that won’t gain you much (if anything?) over just doing the layer swaps on RAM. You can put the text encoder on the second card but you can also just put it in your RAM without much for negatives.
AuryGlenz|6 months ago Nah, that won’t gain you much (if anything?) over just doing the layer swaps on RAM. You can put the text encoder on the second card but you can also just put it in your RAM without much for negatives.
axoltl|7 months ago
karolist|7 months ago
AuryGlenz|6 months ago