-
Notifications
You must be signed in to change notification settings - Fork 4.9k
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
[BUG] DeepSpeed kernel injection crashes on a DeepSeek configuration without rope_theta
bugSomething isn't workingSomething isn't workingStatus: Open.#8340 In deepspeedai/DeepSpeed;[BUG] DeepSpeed accepts negative max_out_tokens and fails generation requests at runtime
bugSomething isn't workingSomething isn't workingStatus: Open.#8339 In deepspeedai/DeepSpeed;[BUG] DeepSpeed generation fails with a CPU/CUDA device mismatch when keep-module-on-host is enabled
bugSomething isn't workingSomething isn't workingStatus: Open.#8338 In deepspeedai/DeepSpeed;[BUG] DeepSpeed kernel injection combined with CUDA graphs aborts at startup
bugSomething isn't workingSomething isn't workingStatus: Open.#8330 In deepspeedai/DeepSpeed;- Status: Open.#8321 In deepspeedai/DeepSpeed;
- Status: Open.#8315 In deepspeedai/DeepSpeed;
- Status: Open.#8291 In deepspeedai/DeepSpeed;
- Status: Open.#8290 In deepspeedai/DeepSpeed;
- Status: Open.#8276 In deepspeedai/DeepSpeed;
Define affine IR for checkpointing
enhancementNew feature or requestNew feature or requestStatus: Open.#8252 In deepspeedai/DeepSpeed;- Status: Open.#8249 In deepspeedai/DeepSpeed;
[BUG]DeepSpeed latest version fails during NCCL process-group initialization with CUDA error 600, while 0.17.2 works
bugSomething isn't workingSomething isn't workingStatus: Open.#8248 In deepspeedai/DeepSpeed;