Search results

Integrate SonicMoE kernels

enhancementNew feature or requestNew feature or request

Status: Open.

#1001 In NVIDIA-NeMo/Automodel;

· hemildesai opened on Dec 29, 2025

Checkpointing on Databricks

bugSomething isn't workingSomething isn't working

Status: Open.

#995 In NVIDIA-NeMo/Automodel;

· jrbourbeau opened on Dec 19, 2025

Extend BaichuanForCausalLM to support seq_lens

enhancementNew feature or requestNew feature or request

Status: Open.

Feature

#994 In NVIDIA-NeMo/Automodel;

· akoumpa opened on Dec 19, 2025

·  

26.02

NVFP4 support

enhancementNew feature or requestNew feature or request

Status: Open.

Feature

#989 In NVIDIA-NeMo/Automodel;

· akoumpa opened on Dec 18, 2025

·  

26.02

CI: Extend test coverage to include downstream customers

enhancementNew feature or requestNew feature or request

Status: Open.

Feature

#980 In NVIDIA-NeMo/Automodel;

· akoumpa opened on Dec 17, 2025

·  

26.02

deepseek-moe-16-base fails on hellaswag dataset

bugSomething isn't workingSomething isn't working

Status: Open.

#972 In NVIDIA-NeMo/Automodel;

· torsli opened on Dec 16, 2025

best_metric_key cannot be set in config

bugSomething isn't workingSomething isn't working

community-request

x-ibm-research

Status: Open.

#971 In NVIDIA-NeMo/Automodel;

· jbross-ibm-research opened on Dec 16, 2025

·  

26.02

Support Bitnet model

enhancementNew feature or requestNew feature or request

x-tii

Status: Open.

#962 In NVIDIA-NeMo/Automodel;

· snowmanwwg opened on Dec 15, 2025

First clip_grad call takes too long

bugSomething isn't workingSomething isn't working

Status: Open.

Bug

#960 In NVIDIA-NeMo/Automodel;

· akoumpa opened on Dec 13, 2025

·  

26.02

Trust issue running deepseek-moe-16B-base

bugSomething isn't workingSomething isn't working

Status: Open.

Bug

#952 In NVIDIA-NeMo/Automodel;

· torsli opened on Dec 10, 2025

·  

26.02

LLM Pre-training example: <code>model_name</code> is required when loading base model

bugSomething isn't workingSomething isn't working

community-request

external

needs-follow-upIssue needs follow-upIssue needs follow-up

x-open

Status: Open.

#950 In NVIDIA-NeMo/Automodel;

· stefan-it opened on Dec 9, 2025

Finetuned checkpoint generating junk outputs

bugSomething isn't workingSomething isn't working

community-request

external

x-typesense

Status: Open.

#945 In NVIDIA-NeMo/Automodel;

· NITHISHM2410 opened on Dec 9, 2025

·  

26.02