Uh oh!
There was an error while loading. Please reload this page.
- Notifications
You must be signed in to change notification settings - Fork 799
All issues
Issue creation is restricted in this repository
Issues
is:issuestate:open
is:issue state:open
Search results
Runtime check for the FlashAttention version rejects local version identifiers when it shouldn't
bugSomething isn't workingSomething isn't workingStatus: Open.#3334In NVIDIA/TransformerEngine;- Status: Open.#3333In NVIDIA/TransformerEngine;
[Bug] Deferred init issue in
te.pytorch.opsbugSomething isn't workingSomething isn't workingStatus: Open.#3322In NVIDIA/TransformerEngine;[Bug] CUDA graph backward replay illegal memory access after validation with mixed NVFP4/MXFP8 Mamba
Status: Open.#3316In NVIDIA/TransformerEngine;- Status: Open.#3312In NVIDIA/TransformerEngine;
- Status: Open.#3310In NVIDIA/TransformerEngine;
- Status: Open.#3309In NVIDIA/TransformerEngine;
[Bug] Transformer Engine SM120 (Blackwell) Compatibility Issue
bugSomething isn't workingSomething isn't workingStatus: Open.#3299In NVIDIA/TransformerEngine;[Bug] Using
QuantizedTensoron CPU causes aRecursionErrorbugSomething isn't workingSomething isn't workingStatus: Open.#3294In NVIDIA/TransformerEngine;[Bug] FusedAttention THD + learnable softmax backward IMA (cuDNN err 700) on Hopper
bugSomething isn't workingSomething isn't workingStatus: Open.#3249In NVIDIA/TransformerEngine;FP8 DPA future token leakage
bugSomething isn't workingSomething isn't workingStatus: Open.#3248In NVIDIA/TransformerEngine;- Status: Open.#3234In NVIDIA/TransformerEngine;