By 
nvidia
LARGE LANGUAGE MODELSTRAINING AND FINE-TUNING
Megatron Core — Ongoing research training transformer models at scale. Listed under Large language models in the Model / Foundation layer. 2 of 17 comparison attributes are recorded, each with the page it was read from.
Megatron Core (NVIDIA) is a library for large-scale transformer model training, the framework behind many of the largest published LLM training runs.
3 OF 4 SIGNALS HELD · MISSING LICENCE
What it runs on, and what runs on it
2 cited edgesComputed from the stack graph, both directions. Every cited edge links the page it was read from; a vendor-added one links the vendor’s own source — this is the reason to be here rather than on the vendor’s own site.
RUNS ON / INTEGRATES WITH
- OpenAIintegrates with sourceJSON
- Weights & Biasesintegrates with sourceJSON
USED BY
- nothing recorded
Open-source signals
A HOLLOW RING IS UNMEASURED, NOT ZERO · WE DO NOT SUBSTITUTE A PARENT COMPANY’S REPOSITORY FOR A PRODUCT THAT HAS NONE
Price
We hold no price for this product.
Either the pricing is behind a sales call and there is no public page to read, or the last reading passed 21 days and was hidden rather than shown stale. If that changes — or if you publish /.well-known/pricing.json — it appears here with a capture date.
How we read prices →What its categories ask
2 answered · 15 unanswered17 questions declared by Large language models and Training and fine-tuning. Most rows across the directory read like these.
“Each line should contain a text field: { "text" : "Your training text here..." }”2026-10-20
A ROW WITH NO ANSWER IS NOT A ROW WITH A “NO” · A VENDOR MAY ANSWER A BLANK ONE, MARKED VENDOR CLAIMED WITH ITS SOURCE · A DOCUMENTED VALUE IS DISPUTED, NOT OVERWRITTEN, AND STAYS VISIBLE WHILE WE RE-READ THE SOURCE
Practitioner reviews
Nobody has written one yet.
A review here is structured, not a star: hardware, scale band, what broke in production, and whether you would choose it again.
NO PERCENTAGE IS SHOWN UNDER FIVE REVIEWS · NO STAR RATINGS, EVER
More from nvidia
6 other products from nvidia. Listed under the same organisation record — see the company page.
garakLLM vulnerability scannerObservability / EvalsNVIDIAWorld Leader in AI ComputingSilicon / Hardware
NVIDIA NeMo GuardrailsNeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systemsSecurity / Governance
NVIDIA NIMPrebuilt inference microservices for foundation modelsInference / Serving
NVIDIA Triton Inference ServerThe Triton Inference Server provides an optimized cloud and edge inferencing solutionInference / Serving
tensorrt-llmTensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUsInference / ServingAlso in this category
64 products in Large language models. Comparison defaults to within the category — pairing this with a GPU cloud would produce a page that means nothing.
USED BY = LISTED PRODUCTS DEPENDING ON IT, IN THIS DIRECTORY
Frequently asked
What does Megatron Core run on?
What does a blank mean on the Megatron Core page?
Where do the Megatron Core figures come from?
Can Megatron Core change what this page says?
Something here is wrong — how do I fix it?
Is this listing run by nvidia?
Cite this page
A model that quotes us should quote us correctly. The reading and its date sit inside the citation rather than beside it, because a quotation carrying only a URL goes wrong the week the URL changes.
infratoapp, “Megatron Core — Large language models”, read 20 October 2026. Hardware: H100 as published at docs.nvidia.com. https://www.infratoapp.com/products/megatron-lm
What Megatron Core can and cannot change here
Claiming nvidia lets its admins and editors change the tagline, description, logo, docs link and deployment model of every product that org owns — live, labelled as theirs, and logged. Answers to blank questions and dependencies go live too, marked as the owner’s and sourced; a documented answer cannot be overwritten. Everything else on this page is a stated fact with a dispute route: price, licence and OSS figures. Those columns are ungranted at the database level, so they are not a feature we withheld — they are a write that cannot happen. Name, category, website, pricing page and package names can be changed by the owner only through our review. A dispute never deletes a value; it marks it and publishes the correction.