GPTQ vs. GGML vs. Base Models: Speed and VRAM Test for Vicuna-33B on A100 GPUgpus.llm-utils.org·1 pts·tikkun·0
Ways LLMs could be used in existing products to add a tiny bit of conveniencegpus.llm-utils.org·12 pts·tikkun·4
Guides for Running Falcon-40B (Base, Instruct, Uncensored) on GPU Cloudsgpus.llm-utils.org·2 pts·tikkun·1
Show HN: Compute Watch – Track cloud prices and availability of A100s and H100scomputewatch.llm-utils.org·1 pts·tikkun·1
Nvidia H100 and A100 GPUs – comparing available capacity at GPU cloud providersllm-utils.org·183 pts·tikkun·126