Repository Issue Activity (beta)

#325 Feature: Adaptive GPU loading — fill VRAM to maximize speed (full-load bypass v1 PR ready)

#318 AMD ROCM support?

#328 Login button is not working on Homepage

#326 ModuleNotFoundError: Could not import module 'LlamaForCausalLM'. Are this object's requirements defined correctly?

#335 ValueError: invalid literal for int() with base 10: 'layers'

#331 from torch._C import * # noqa: F403

#330 3x speedup compression claim is false/outdated

#301 run this with windows amd 9070 XT ROCm 7.12 VERRY SLOW

#300 买了不少流量吧，墙内墙外天天推，几年不更新的玩意儿你要干啥

#317 Running old GPUs ?

#319 grok claude

#312 Proposal: a lightweight CI + test/benchmark harness

#262 is it only for nvidia (cuda )

#210 How to use CPU inference?

#311 Your project is on StackMap — a curated map of the AI stack

#304 Using GGUF

#277 Cant install requements.txt

#44 More clever batching of layers

#299 can this project run deepseek V4 flash on an 8G VRam?

#284 I found the perfect use case for airllm

#198 unsloth/Llama-3.1-Nemotron-70B-Instruct-bnb-4bit

#180 unsloth/Meta-Llama-3.1-405B-Instruct-bnb-4bit

#146 Add support for Mistral model inference

#154 how to use Qwen2-72B-instuct

#285 Any plan to support for qwen3.6-35b-a3b?