docs(ultra3): update quantization recipe and precision table#257
Merged
marcromeyn merged 1 commit intoJun 5, 2026
Merged
Conversation
Replace the generic NVFP4 description with the actual hybrid recipe used for the Nemotron 3 Ultra release: NVFP4 W4A4 for routed experts, FP8 per-tensor for shared experts and Mamba mixer linears, FP8 KV cache, BF16 for attention / latent MoE / lm_head / MTP / embeddings. - Rename "NVFP4 Checkpoint" → "NVFP4 + FP8 Hybrid Checkpoint" - Update the precision table to match super-nvfp4-max-calib.yaml - Drop the obsolete "Quantization Configurations" four-name table (the released recipe is a YAML file, not a built-in shorthand name) - Switch the Quantize command to point at the recipe YAML via --export-quant-cfg - Tighten memory-reduction claim to the measured ratio (~3.3×, 1.1 TB → 331 GB) and parameter count (≈560B) - Rename "Recipe Execution" → "Commands" and drop the Direct-Script-Execution preamble (redundant with Megatron-Bridge README) - Drop the "Parallelism Configuration" subsection (already captured by the --pp/--tp/--ep flags shown in each command) Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> Signed-off-by: James Shen <yueshen@nvidia.com>
26329a8 to
4d1b27b
Compare
marcromeyn
approved these changes
Jun 5, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Replace the generic NVFP4 description in
docs/nemotron/ultra3/quantization.mdwith the actual hybrid recipe used for the released Nemotron 3 Ultra checkpoint, and align the precision settings table withsuper-nvfp4-max-calib.yaml.Changes
mixer.{in,out}_proj, FP8 KV cache, BF16 everywhere else (attention, latent MoE, MTP, lm_head, embeddings, Mamba conv1d/SSM).Recipe Executionsection. Updates the quantize command to pass the YAML recipe path via--export-quant-cfginstead of the built-inmamba_moe_nvfp4_conservativeshorthand (the released recipe is a YAML file, not a built-in name).~4×to the measured~3.3×(1.1 TB BF16 → 331 GB hybrid).--pp/--tp/--epflags shown in each command.Test plan
*mlp.experts*→ NVFP4,*mlp.shared_experts*and*mixer.{in,out}_proj*→ FP8, everything else disabled).examples/quantization/{quantize,ptq_generate,export}.pyin Megatron-Bridge.super-nvfp4-max-calib.yamlrecipe path and that PTQ + export + sanity-prompt round-trip works end-to-end on this checkpoint.🤖 Generated with Claude Code