-
Notifications
You must be signed in to change notification settings - Fork 613
All issues
Issue creation is restricted in this repository
- #1699 · Trenton-Starkey opened
on Jun 12, 2026 4
Issues
is:issue state:open
is:issue state:open
Search results
Handling vocabulary size mismatches between teacher and student in distillation
feature requestNew feature or requestNew feature or requestStatus: Open.#2489 In NVIDIA/Model-Optimizer;StaticLossBalancer warns "weights do not sum to 1.0" with a config that does sum to 1.0
bugSomething isn't workingSomething isn't workingStatus: Open.#2488 In NVIDIA/Model-Optimizer;Repeated "intermediate output already stored" warnings during gradient accumulation
bugSomething isn't workingSomething isn't workingStatus: Open.#2487 In NVIDIA/Model-Optimizer;[Puzzletron] KV-cache memory computes head size as n_embd // n_head instead of reading head_dim
bugSomething isn't workingSomething isn't workingStatus: Open.#2482 In NVIDIA/Model-Optimizer;- Status: Open.#2481 In NVIDIA/Model-Optimizer;
qdq_to_dq fails on MaxViT due to unsupported Transpose after DequantizeLinear
bugSomething isn't workingSomething isn't workingStatus: Open.#2442 In NVIDIA/Model-Optimizer;calling mtq.quantize causes AttributeError: 'tuple' object has no attribute 'dim'
questionHelp is is neededHelp is is neededStatus: Open.#2424 In NVIDIA/Model-Optimizer;Only 1 sample from AutoCast calibration data set is truly used
bugSomething isn't workingSomething isn't workingStatus: Open.#2337 In NVIDIA/Model-Optimizer;After applying NVFP4 quantization followed by compression, RealQuantLinear fails to locate the corresponding GEMM implementation during inference.
bugSomething isn't workingSomething isn't workingStatus: Open.#2330 In NVIDIA/Model-Optimizer;- Status: Open.#2281 In NVIDIA/Model-Optimizer;
hf_ptq.py: deprecated --auto_quantize_bits CLI flag silently no-ops instead of enabling AutoQuantize
Status: Open.#2230 In NVIDIA/Model-Optimizer;[Feature Request] DeepSeek-V4-Flash-0731 NVFP4 checkpoint
feature requestNew feature or requestNew feature or requestStatus: Open.#2220 In NVIDIA/Model-Optimizer;