Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →nvidia/nemo-automodel-model-onboarding
Guide for onboarding new model architectures into NeMo AutoModel, including architecture discovery, implementation patterns, registration, and validation.
Passed 22 of 23 test cases with the skill, and 4 of 23 without it.
and 27 more in the full report
18 fixed4 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 18 test cases the skill fixed. Same model, same prompt, with and without the skill.
What are the exact enum values accepted by NeMo AutoModel for `tie_word_embeddings_support` on registered causal language models?
| Checked | Without skill | With skill |
|---|---|---|
| Lists BOTH, TIED_ONLY, and UNTIED_ONLY as the valid tie support options. | Missed | Passed |
The instructions your model reads when the skill loads.
This skill guides implementation of new model architectures in NeMo AutoModel. Follow the five phases in order.
When answering an onboarding question, keep the response in this order:
config.json.components/models/<name>/.For conceptual onboarding questions, answer from this skill without opening the pattern files unless the user asks you to edit code. Mention pattern filenames as references, then give the direct checklist.
Use direct action verbs: classify the model, name the files, map the weights, register the class, and add tests. Do not discuss distributed strategy, launcher configuration, or general recipe authoring unless the user explicitly connects it to onboarding a new architecture.
Use these compact answer patterns for common questions:
architectures contains a
ForCausalLM class and expert fields such as num_local_experts,
n_routed_experts, or num_experts_per_tok are absent. Create
components/models/<name>/model.py, state_dict_adapter.py, __init__.py,
and optional config.py, register MODEL_ARCH_MAPPING in
_transformers/registry.py, add example YAML, and add tiny-config unit tests
plus layer-equivalence tests for rewritten layers.config.json, reference
moe-patterns.md, map router tensors separately, preserve routed-expert
index order, map routed experts, shared experts, and gate/up/down projections,
add adapter key-map tests and tiny-config numerical equivalence tests, and dotest cases passed, vs 4 / 23 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.