Specification

Release date: 1 May 2026
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Openness: open-weights
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
License: apache-2.0
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Architecture: Gemma4ForConditionalGeneration
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Parameters: 14.4B
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
Active parameters: 4B
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Base model: google/gemma-4-26B-A4B-it
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
File size: 18.8 GB
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Hugging Face repo: nvidia/Gemma-4-26B-A4B-NVFP4
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
Pipeline tag: text-generation
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
Model card: huggingface.co/nvidia/Gemma-4-26B-A4B-NVFP4
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
Downloads: 1,850,407
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Likes: 138
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Gated: No
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Last modified: 2026-05-11T15:58:52+00:00
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 h agomedium
Library name: Model Optimizer
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
License url: https://ai.google.dev/gemma/apache_2
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Downloads all time: 7,434,326
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Model type: gemma4
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Tags: Model Optimizer, safetensors, gemma4, nvidia, ModelOpt, quantized, NVFP4, nvfp4, gemma4-26b-A4B-it, text-generation, 8-bit, modelopt
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Weights dtype: BF16, F8_E4M3, U8
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium

Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →

Provenance

Attributed facts

21

Source tiers

T221

Freshest observation

1 h ago

Conflicts

None

Modalities

Modalities unavailable.

Capabilities

Tool calling
Unavailable
Structured output
Unavailable
Reasoning
Unavailable
Vision
Unavailable
Audio
Unavailable
Fine-tuning available
Unavailable

No capability flags have been observed from a source yet — we do not infer them.

No structured attributes yet.

No benchmark results recorded

Results appear when a tier 1–3 source publishes them; we never copy scores without a source.

Current prices

No current prices recorded

Prices appear when a provider publishes a public pricing page we crawl.

Price history

Estimated

Memory need = bytes per parameter (4-bit ≈ 0.5 × 1.15 overhead, 8-bit 1.0, fp16 2.0) + a KV-cache allowance. Not a measurement.

Estimated hardware fit
Hardware	Quantization	Memory	Est. need	Fits
Apple M3 Ultra	4bit	—	8.8 GB	Yes
AMD Instinct MI325X	4bit	256 GB	8.8 GB	Yes
AMD Instinct MI300X	4bit	192 GB	8.8 GB	Yes
Apple M2 Ultra	4bit	—	8.8 GB	Yes
NVIDIA B200	4bit	180 GB	8.8 GB	Yes
NVIDIA H200	4bit	141 GB	8.8 GB	Yes
Apple M1 Ultra	4bit	—	8.8 GB	Yes
Apple M3 Max	4bit	—	8.8 GB	Yes
Apple M4 Max	4bit	—	8.8 GB	Yes
NVIDIA DGX Spark	4bit	128 GB	8.8 GB	Yes
Apple M2 Max	4bit	—	8.8 GB	Yes
NVIDIA A100 80GB	4bit	80 GB	8.8 GB	Yes
NVIDIA H100 SXM	4bit	80 GB	8.8 GB	Yes
Apple M1 Max	4bit	—	8.8 GB	Yes
Apple M4 Pro	4bit	—	8.8 GB	Yes
Apple M3 Pro	4bit	—	8.8 GB	Yes
Apple M1 Pro	4bit	—	8.8 GB	Yes
Apple M2 Pro	4bit	—	8.8 GB	Yes
Apple M4	4bit	—	8.8 GB	Yes
NVIDIA GeForce RTX 5090	4bit	32 GB	8.8 GB	Yes
Apple M2	4bit	—	8.8 GB	Yes
Apple M3	4bit	—	8.8 GB	Yes
NVIDIA GeForce RTX 3090	4bit	24 GB	8.8 GB	Yes
NVIDIA GeForce RTX 4090	4bit	24 GB	8.8 GB	Yes
Apple M1	4bit	—	8.8 GB	Yes

Ancestors 2

Gemma 4 26B A4B25.8B
gemma-4-26B-A4BGoogle

This model

nvidia/Gemma-4-26B-A4B-NVFP4

14.4B params

Quantizations 0

None recorded.

Descendants 0

None recorded.

Papers 0

No papers linked yet.

Repositories 0

No repositories linked yet.

As of

Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.

Claim history · Model card

1 claims · 1 propertiesShow all properties

Model cardmodel_card_url1

Claim history for Model card
Value	Valid from → to	Status	Source	Confidence	Extractor
https://huggingface.co/nvidia/Gemma-4-26B-A4B-NVFP4	11 Sept 2026 21:11 UTC → current	current	Hugging Face Hub (public pages, model cards, papers)T2	medium	deterministic

Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →

11 Sept 2026
New modelModelnvidia/Gemma-4-26B-A4B-NVFP4NVIDIA
New model: nvidia/Gemma-4-26B-A4B-NVFP4 (NVIDIA)
huggingface
huggingface.co

Full timeline →

Source documents
Source	Document	Type	Tier	Last observed	Snapshots
Hugging Face Hub (public pages, model cards, papers)	huggingface.co/nvidia/Gemma-4-26B-A4B-NVFP4	model_page	T2· Quality secondary	1 h ago	1
Hugging Face Hub (public pages, model cards, papers)	huggingface.co/models?author=nvidia&p=0&sort=downloads	listing	T2· Quality secondary	2 h ago	2

Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.