Search results

Does the SFT pipeline support LoRA / PEFT for training on lower-VRAM setups?

Status: Open.

#106 In NVlabs/alpamayo;

· minji00116-ctrl opened on May 8, 2026

Duplicate preprocessing in dataset

Status: Open.

#105 In NVlabs/alpamayo;

· clmforgit opened on May 7, 2026

About the weight provided in this release after which training stage

Status: Open.

#71 In NVlabs/alpamayo;

· DFLyan opened on Apr 21, 2026

Feature suggestion

Status: Open.

#70 In NVlabs/alpamayo;

· arthurshelby3210-sudo opened on Apr 20, 2026

Current limitations

Status: Open.

#69 In NVlabs/alpamayo;

· arthurshelby3210-sudo opened on Apr 19, 2026

Question about Alpamayo SFT CoC

Status: Open.

#68 In NVlabs/alpamayo;

· moonzerokevin opened on Apr 19, 2026

Question about RL post-training for the action expert pathway

Status: Open.

#67 In NVlabs/alpamayo;

· RexGPT opened on Apr 9, 2026

why use image input not video input in finetune?

Status: Open.

#65 In NVlabs/alpamayo;

· yaogang2060 opened on Apr 7, 2026

List index out of range while loading the dataset

Status: Open.

#61 In NVlabs/alpamayo;

· ParvezAlam123 opened on Mar 25, 2026

An error occurred when running test_inference.py

Status: Open.

#59 In NVlabs/alpamayo;

· yaoyannan1111 opened on Mar 25, 2026

Evaluating in the closed-loop simulator HUGSIM

Status: Open.

#54 In NVlabs/alpamayo;

· TheNeeloy opened on Mar 13, 2026

Enabling flash_attention_2 causes RuntimeError: cu_seqlens_q must have shape (batch_size + 1)

Status: Open.

#52 In NVlabs/alpamayo;

· hyc-hw opened on Mar 2, 2026