Llama-3.1-Nemotron-70B-Instruct-HF VRAM & Memory Requirements

Specific checkpoints with sourced architecture and weight files. GGUF Q4_K_M (bartowski): 42.23 GiB–43.57 GiB at 8,192 tokens, 1 sequence, FP16 KV, llama.cpp.

This result is an estimate, not a measured trace.Not verifiedFile-backed weights

Runs on

Can Llama-3.1-Nemotron-70B-Instruct-HF run on Apple M2 Max?

Publisher and provenance

PublisherNVIDIA
Hugging Facenvidia/Llama-3.1-Nemotron-70B-Instruct-HF
Revision031d4042f36adc1a52cca51b331d25cbe3cf1022
Total parameters70,553,706,496
Layers80
Query heads64
KV heads8 (gqa)
Head dimension128
Recorded context capability131,072
Completenesscomplete · full

Memory layers in this estimate

Artifact bytes, raw KV, engine extras, and total demand are separate. This is not a runtime trace.

Weight artifact: 39.60 GiB (42,520,398,752 bytes)
Raw KV cache: 2.50 GiB (2,684,354,560 bytes)
Engine extras / workspace: 0.13 GiB–1.47 GiB
Total estimated demand: 42.23 GiB–43.57 GiB

Published artifacts

Quantization	Selected files	Bytes	Engine	Evidence
BF16 safetensors bf16 Calculator: BF16 safetensors	30 safetensors shards Total selected bytes: 141,107,412,992 (131.42 GiB / 141.11 GB) First shard: model-00001-of-00030.safetensors Last shard: model-00030-of-00030.safetensors View all 30 shard files `model-00001-of-00030.safetensors` `model-00002-of-00030.safetensors` `model-00003-of-00030.safetensors` `model-00004-of-00030.safetensors` `model-00005-of-00030.safetensors` `model-00006-of-00030.safetensors` `model-00007-of-00030.safetensors` `model-00008-of-00030.safetensors` `model-00009-of-00030.safetensors` `model-00010-of-00030.safetensors` `model-00011-of-00030.safetensors` `model-00012-of-00030.safetensors` `model-00013-of-00030.safetensors` `model-00014-of-00030.safetensors` `model-00015-of-00030.safetensors` `model-00016-of-00030.safetensors` `model-00017-of-00030.safetensors` `model-00018-of-00030.safetensors` `model-00019-of-00030.safetensors` `model-00020-of-00030.safetensors` `model-00021-of-00030.safetensors` `model-00022-of-00030.safetensors` `model-00023-of-00030.safetensors` `model-00024-of-00030.safetensors` `model-00025-of-00030.safetensors` `model-00026-of-00030.safetensors` `model-00027-of-00030.safetensors` `model-00028-of-00030.safetensors` `model-00029-of-00030.safetensors` `model-00030-of-00030.safetensors`	141,107,412,992 (131.42 GiB / 141.11 GB)	`vllm, sglang, unspecified`	`file-metadata`
GGUF Q4_K_M (bartowski) gguf-q4_k_m Calculator: GGUF Q4_K_M (bartowski)	Llama-3.1-Nemotron-70B-Instruct-HF-Q4_K_M.gguf	42,520,398,752 (39.60 GiB / 42.52 GB)	`llama.cpp, ollama, unspecified`	`file-metadata`
GGUF Q5_K_M (bartowski) gguf-q5_k_m Calculator: GGUF Q5_K_M (bartowski)	Llama-3.1-Nemotron-70B-Instruct-HF-Q5_K_M/Llama-3.1-Nemotron-70B-Instruct-HF-Q5_K_M-00001-of-00002.gguf, Llama-3.1-Nemotron-70B-Instruct-HF-Q5_K_M/Llama-3.1-Nemotron-70B-Instruct-HF-Q5_K_M-00002-of-00002.gguf	49,949,822,016 (46.52 GiB / 49.95 GB)	`llama.cpp, ollama, unspecified`	`file-metadata`
GGUF Q6_K (bartowski) gguf-q6_k Calculator: GGUF Q6_K (bartowski)	Llama-3.1-Nemotron-70B-Instruct-HF-Q6_K/Llama-3.1-Nemotron-70B-Instruct-HF-Q6_K-00001-of-00002.gguf, Llama-3.1-Nemotron-70B-Instruct-HF-Q6_K/Llama-3.1-Nemotron-70B-Instruct-HF-Q6_K-00002-of-00002.gguf	57,888,148,576 (53.91 GiB / 57.89 GB)	`llama.cpp, ollama, unspecified`	`file-metadata`
GGUF Q8_0 (bartowski) gguf-q8_0 Calculator: GGUF Q8_0 (bartowski)	Llama-3.1-Nemotron-70B-Instruct-HF-Q8_0/Llama-3.1-Nemotron-70B-Instruct-HF-Q8_0-00001-of-00002.gguf, Llama-3.1-Nemotron-70B-Instruct-HF-Q8_0/Llama-3.1-Nemotron-70B-Instruct-HF-Q8_0-00002-of-00002.gguf	74,975,054,944 (69.83 GiB / 74.98 GB)	`llama.cpp, ollama, unspecified`	`file-metadata`
INT4 theoretical	No published file	35,276,853,248 (theoretical)	—	theoretical

Memory by quantization and context

Concurrent sequences = 1. KV defaults to FP16. The range includes documented runtime extras for the listed engine. These figures use the same core as the interactive calculator.

Quantization	2,048 tokens	8,192 tokens	32,768 tokens
BF16 safetensors	132.67 GiB–134.56 GiB	134.54 GiB–136.49 GiB	142.04 GiB–144.22 GiB
GGUF Q4_K_M (bartowski)	40.35 GiB–41.53 GiB	42.23 GiB–43.57 GiB	49.73 GiB–51.59 GiB
GGUF Q5_K_M (bartowski)	47.27 GiB–48.59 GiB	49.14 GiB–50.62 GiB	56.64 GiB–58.65 GiB
GGUF Q6_K (bartowski)	54.66 GiB–56.13 GiB	56.54 GiB–58.17 GiB	64.04 GiB–66.19 GiB
GGUF Q8_0 (bartowski)	70.58 GiB–72.36 GiB	72.45 GiB–74.40 GiB	79.95 GiB–82.42 GiB
INT4 theoretical Theoretical rows are not downloadable artifacts.	33.60 GiB–43.62 GiB	35.48 GiB–45.97 GiB	42.98 GiB–55.20 GiB

Raw KV contribution

Standard GQA/MHA raw KV is 2 × layers × kvHeads × headDim × tokens × sequences × bytes per element. This page uses KV heads (8), not query heads (64). Using query heads would overstate this cache by 8×.

8,192 tokens, 1 sequence, FP16: 2,684,354,560 bytes (2.5000 GiB)
32,768 tokens: 10,737,418,240 bytes (10.00 GiB)

Open the calculator with this model

Sources

Llama-3.1-Nemotron-70B-Instruct-HF model card
config.json — revision 031d4042f36adc1a52cca51b331d25cbe3cf1022
safetensors index — metadata.total_size=141107412992
Artifact repo bartowski/Llama-3.1-Nemotron-70B-Instruct-HF-GGUF

Assumptions that affect this result

Requested context is within the recorded native window of 131072 tokens.
Full-resident weight estimates use total parameters, including inactive MoE experts.
Weight bytes come from the selected weight files only, not from every file in the repository.
Download file size is not peak GPU memory. Runtime layout, allocator padding, and KV are extra.
Counted files: Llama-3.1-Nemotron-70B-Instruct-HF-Q4_K_M.gguf.
Context budget is the number of cached tokens per sequence (prompt plus reserved generation).
Concurrent sequences share one model replica; weights are not multiplied by sequence count.
KV dtype is independent of weight quantization unless you change it.

Calculator

Loading catalog snapshot…

Choose a model first, then check a device.

Model

Show historical models

Quantization

Device type

Hardware

Workload

Context length primarily affects KV cache memory.

Advanced settings

Runtime

Engine and KV cache type for this estimate.

EngineKV cache typeRecurrent / conv state dtype Include theoretical estimates (expert) Allow a substitute if the filter has no match

Workload

Overhead scenario and the prefill workspace.

Overhead scenario Include heavier prefill workspace

Topology

How the model is split across devices. Memory is not added together.

DeploymentIntra-replica splitDevice count

Replica count is an advanced deployment option. Quantity is nodes or GPUs, not replicas.

Replica groupsDevices per replica

Budget

System reserve and safety margin. An empty number keeps the previous value.

System / display reserve

GiB (2^30 bytes)

Budget already excludes system reserveExtra safety margin

Custom bits per weightDisplay unit

Custom model

Import a config or type the architecture. Nothing is scanned until you import.

Import local config.json

Parsed in the browser. The file is not uploaded.

Public Hugging Face repoRevision

Only huggingface.co, config.json, no token. Private repos are rejected.

Total parametersActive parametersLayersQuery headsKV headsHead dimension

Search or pick a catalog checkpoint to see memory demand.

Search or pick a catalog checkpoint to see memory demand.