Skip to content
archdiff

models / compare / nemotron-3-ultra-vs-minimax-m3

Nemotron 3 Ultra vs MiniMax-M3

How Nemotron 3 Ultra (NVIDIA, June 2026) and MiniMax-M3 (MiniMax, June 2026) are built, drawn on the same November 2026 blueprint. They make different choices on 11 of 27 components. Select a component in either diagram to find it in the table.

--- frontier-llm@2026-11+++ nemotron-3-ultra@2026-06+0~4−7

Nemotron 3 Ultra

NVIDIA, June 2026, 550B total, 55B active

Text · code · tool outputText · code · tool outputImages · video · audioImages · video · audioByte-level BPE tokenizer — 128K–262K vocabByte-level BPE tokenizer128K–262K vocabVision / audio encoder — native ViT or encoder-free patchesVision / audio encodernative ViT or encoder-free patchesToken embedding — media embeds spliced in placeToken embeddingmedia embeds spliced in placeEngram n-gram memory — hashed 2–4-gram lookup (V4.1)Engram n-gram memoryhashed 2–4-gram lookup (V4.1)Single residual stream, one sublayer per layer — Residual & normSingle residual stream, one sublayer p…Residual & normRMSNorm — pre-normRMSNormpre-normQK-Norm — per-head RMSNorm on q, kQK-Normper-head RMSNorm on q, kPartial RoPE / NoPE — rotate 64 dims, or nonePartial RoPE / NoPErotate 64 dims, or noneMamba-2 selective state-space layers — AttentionMamba-2 selective state-spac…AttentionGlobal GQA attention without positional encoding — AttentionGlobal GQA attention without…AttentionSparse / compressed attn — DSA indexer · CSA / HCASparse / compressed attnDSA indexer · CSA / HCASliding-window attention — local · 128–1,024-token windowSliding-window attentionlocal · 128–1,024-token windowOutput gate + sink — σ(x·W_g) ⊙ o · learnable sinkOutput gate + sinkσ(x·W_g) ⊙ o · learnable sink⊕ residual⊕residualRMSNorm — pre-normRMSNormpre-normMoE router — σ / √softplus · bias balancingMoE routerσ / √softplus · bias balancingLatentMoE routed experts (ReLU², 2,048-dim latent) — Mixture of expertsLatentMoE routed experts (ReLU², 2,048…Mixture of expertsShared expert — 1–2 · always activeShared expert1–2 · always active⊕ residual⊕residualFinal RMSNormFinal RMSNormLM head — untied · vocab logitsLM headuntied · vocab logitsMTP module(s) — predict t+2 … t+kMTP module(s)predict t+2 … t+kNext-token logitsNext-token logitsDraft tokensDraft tokensMuon optimizer — Newton–Schulz · QK-Clip · per-headMuon optimizerNewton–Schulz · QK-Clip · per-headFP8 → FP4 training — block-scaled FP8 · MXFP4/NVFP4 QATFP8 → FP4 trainingblock-scaled FP8 · MXFP4/NVFP4 QATLong-context extension — staged 4K → 1M · YaRNLong-context extensionstaged 4K → 1M · YaRNRL with verifiable rewards — GRPO family · async agentic RLRL with verifiable rewardsGRPO family · async agentic RLOn-policy distillation — reverse KL to specialist teachersOn-policy distillationreverse KL to specialist teachersSpeculative decoding — MTP self-drafting · DSparkSpeculative decodingMTP self-drafting · DSparkKV-cache compression — FP4 KV · cross-layer reuse · SSDKV-cache compressionFP4 KV · cross-layer reuse · SSD
AttentionMixture of expertsResidual & normInput & outputTraining & inferenceData flowResidual / skipAddedReplacedNot used

--- frontier-llm@2026-11+++ minimax-m3@2026-06+0~4−4

MiniMax-M3

MiniMax, June 2026, 427.0B total, 23.5B active

Text · code · tool outputText · code · tool outputImages · video · audioImages · video · audioByte-level BPE tokenizer — 128K–262K vocabByte-level BPE tokenizer128K–262K vocabVision / audio encoder — native ViT or encoder-free patchesVision / audio encodernative ViT or encoder-free patchesToken embedding — media embeds spliced in placeToken embeddingmedia embeds spliced in placeEngram n-gram memory — hashed 2–4-gram lookup (V4.1)Engram n-gram memoryhashed 2–4-gram lookup (V4.1)Single residual stream — Residual & normSingle residual streamResidual & normRMSNorm — pre-normRMSNormpre-normQK-Norm — per-head RMSNorm on q, kQK-Normper-head RMSNorm on q, kPartial RoPE / NoPE — rotate 64 dims, or nonePartial RoPE / NoPErotate 64 dims, or noneGated DeltaNet / KDA — linear · ~3 of every 4 layersGated DeltaNet / KDAlinear · ~3 of every 4 layersFull grouped-query attention in blocks 0–2 — AttentionFull grouped-query attention…AttentionMiniMax Sparse Attention (MSA) — AttentionMiniMax Sparse Attention (MS…AttentionSliding-window attention — local · 128–1,024-token windowSliding-window attentionlocal · 128–1,024-token windowOutput gate + sink — σ(x·W_g) ⊙ o · learnable sinkOutput gate + sinkσ(x·W_g) ⊙ o · learnable sink⊕ residual⊕residualRMSNorm — pre-normRMSNormpre-normMoE router — σ / √softplus · bias balancingMoE routerσ / √softplus · bias balancingRouted experts — top-6–22 of 256–896 · SwiGLURouted expertstop-6–22 of 256–896 · SwiGLUShared expert — 1–2 · always activeShared expert1–2 · always active⊕ residual⊕residualFinal RMSNormFinal RMSNormLM head — untied · vocab logitsLM headuntied · vocab logitsMTP module(s) — predict t+2 … t+kMTP module(s)predict t+2 … t+kNext-token logitsNext-token logitsDraft tokensDraft tokensMuon optimizer — Newton–Schulz · QK-Clip · per-headMuon optimizerNewton–Schulz · QK-Clip · per-headBF16 weights plus an official MXFP8 checkpoint — Training & inferenceBF16 weights plus an official MXFP8 ch…Training & inferenceLong-context extension — staged 4K → 1M · YaRNLong-context extensionstaged 4K → 1M · YaRNRL with verifiable rewards — GRPO family · async agentic RLRL with verifiable rewardsGRPO family · async agentic RLOn-policy distillation — reverse KL to specialist teachersOn-policy distillationreverse KL to specialist teachersSpeculative decoding — MTP self-drafting · DSparkSpeculative decodingMTP self-drafting · DSparkKV-cache compression — FP4 KV · cross-layer reuse · SSDKV-cache compressionFP4 KV · cross-layer reuse · SSD
AttentionMixture of expertsResidual & normInput & outputTraining & inferenceData flowResidual / skipAddedReplacedNot usedNot disclosed

Component by component

Nemotron 3 Ultra and MiniMax-M3 differ on 11 of 27 components.

Blueprint componentNemotron 3 UltraMiniMax-M3

Not used

Same as the blueprintA 32-layer CLIP-style ViT (width 1,280, 16 heads, GELU MLP of 5,120, LayerNorm; 631M parameters) embeds 14 × 14 patches with a Conv3d that spans 2 frames for video, using 3D (time, height, width) RoPE. A GELU MLP lifts each patch to 6,144 and a second fuses each 2 × 2 group into one token (234M projector parameters). MiniMax says M3 trained on mixed modalities from the first step and lists text, image and video inputs.

Single residual stream, one sublayer per layer108 pre-norm layers on one d = 8,192 residual stream; each layer is one mixer or one MoE, not an attention+FFN pair.

Single residual streamOne d = 6,144 residual stream with pre-norm sublayers; no hyper-connections.

Not used

Same as the blueprintPer-head Gemma-style RMSNorm (ε = 1e-6) on q and k before RoPE, with one 128-dim gain shared by all 64 query heads and one by the 4 key heads; the indexer's queries and key get their own. This replaces M2's layer-wide norm over the concatenated projections (qk_norm_type per_layer → per_head), so head-sharded tensor parallelism no longer needs a cross-rank reduction for it.

Mamba-2 selective state-space layers48 of the 60 mixer layers: SSD with 256 heads × 64 dims, state size 128, 8 B/C groups, causal conv of kernel 4.

Not used

Global GQA attention without positional encoding12 layers of dense causal attention: 64 query heads share 2 KV heads (d_h = 128); no RoPE, QK-norm, gate or sink.

Full grouped-query attention in blocks 0–2The first three blocks attend over the whole context: 64 query heads share 4 KV heads of dim 128; no indexer, gate or sink.

Not used

MiniMax Sparse Attention (MSA)Per GQA group, a 2-projection indexer picks the top 16 of 128-token KV blocks; exact softmax over ≤ 2,048 keys, in 57 of 60 layers.

LatentMoE routed experts (ReLU², 2,048-dim latent)512 non-gated squared-ReLU experts of width 5,120 run in a 2,048-dim latent space; 22 active per token.

Same as the blueprint128 experts of intermediate size 3,072, top-4 active, in blocks 3–59 (M2: 256 of width 1,536, top-8). Blocks 0–2 use a dense MLP of width 12,288 instead. Every FFN uses gpt-oss's clamped SwiGLU: gate clamped at 7, linear branch to ±7, output (x_lin + 1) · x_glu · σ(1.702 · x_glu). Routed experts hold 413.1B of the 427.0B parameters.

Not used

Not disclosed

Same as the blueprintPretrained on 20T tokens in NVFP4 (E2M1 with 16-element blocks, 2D weight scaling, Random Hadamard Transforms on wgrad inputs, stochastic rounding of gradients). The final 16 layers, Mamba output projections, latent projections, QKV/attention projections, MTP layers and embeddings stayed in higher precision. The train-loss gap to BF16 branches stayed below 0.4% on average.

BF16 weights plus an official MXFP8 checkpointBF16 release with an FP32 router; MiniMax-M3-MXFP8 stores the text stack's linears as E4M3 with one 8-bit scale per 32 weights.

Same as the blueprintMulti-teacher on-policy distillation (MOPD) in two iterations. More than ten domain-specialized teachers (STEM, coding, SWE, search, terminal use, office work, chat, safety and others) give dense token-level reverse-KL signals on the student's own rollouts. The RLVR student acts as a self-teacher for uncovered domains.

Not disclosed

Same as the blueprintThe shared MTP head drafts recursively; the published serving recipes default to 5 draft tokens. On SPEED-Bench at draft length 7 with greedy decoding, the mean acceptance length is 4.39 tokens, and 4.58 after MTP Boosting.

Not disclosed

More comparisons