Specification

Release date: 6 Dec 2024
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
Openness: open-weights
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
License: other
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
Architecture: LlamaForCausalLM
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
Parameters: 70.5B
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Base model: meta-llama/Llama-3.3-70B-Instruct
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
Quantization format: fp8
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
File size: 72.7 GB
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
Hugging Face repo: amd/Llama-3.3-70B-Instruct-FP8-KV
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Model card: huggingface.co/amd/Llama-3.3-70B-Instruct-FP8-KV
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
Downloads: 13,748
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 54 min agomedium
Likes: 4
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 54 min agomedium
Gated: No
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
Is quantized: Yes
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
Last modified: 2025-07-01T09:32:05+00:00
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 1 h agomedium
License name: llama3.3
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 2 min agomedium
License url: https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/LICENSE
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
Downloads all time: 218,573
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 54 min agomedium
Model type: llama
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
Tags: safetensors, llama, fp8
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min agomedium
Weights dtype: BF16, F8_E4M3
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 58 min 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

2 min 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	—	41.1 GB	Yes
AMD Instinct MI325X	4bit	256 GB	41.1 GB	Yes
AMD Instinct MI300X	4bit	192 GB	41.1 GB	Yes
Apple M2 Ultra	4bit	—	41.1 GB	Yes
NVIDIA B200	4bit	180 GB	41.1 GB	Yes
NVIDIA H200	4bit	141 GB	41.1 GB	Yes
Apple M1 Ultra	4bit	—	41.1 GB	Yes
Apple M3 Max	4bit	—	41.1 GB	Yes
Apple M4 Max	4bit	—	41.1 GB	Yes
NVIDIA DGX Spark	4bit	128 GB	41.1 GB	Yes
Apple M2 Max	4bit	—	41.1 GB	Yes
NVIDIA A100 80GB	4bit	80 GB	41.1 GB	Yes
NVIDIA H100 SXM	4bit	80 GB	41.1 GB	Yes
Apple M1 Max	4bit	—	41.1 GB	Yes
Apple M4 Pro	4bit	—	41.1 GB	Yes
Apple M3 Pro	4bit	—	41.1 GB	No
Apple M1 Pro	4bit	—	41.1 GB	No
Apple M2 Pro	4bit	—	41.1 GB	No
Apple M4	4bit	—	41.1 GB	No
NVIDIA GeForce RTX 5090	4bit	32 GB	41.1 GB	No
Apple M2	4bit	—	41.1 GB	No
Apple M3	4bit	—	41.1 GB	No
NVIDIA GeForce RTX 3090	4bit	24 GB	41.1 GB	No
NVIDIA GeForce RTX 4090	4bit	24 GB	41.1 GB	No
Apple M1	4bit	—	41.1 GB	No

Ancestors 2

Llama 3.3 70B Instruct70.5B
Llama-3.1-70BMeta AI

This model

amd/Llama-3.3-70B-Instruct-FP8-KV

70.5B 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 · Weights dtype

1 claims · 1 propertiesShow all properties

Weights dtypeweights_dtype1

Claim history for Weights dtype
Value	Valid from → to	Status	Source	Confidence	Extractor
BF16, F8_E4M3	11 Sept 2026 21:15 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 modelModelamd/Llama-3.3-70B-Instruct-FP8-KVAMD
New model: amd/Llama-3.3-70B-Instruct-FP8-KV (AMD)
huggingface
huggingface.co

Full timeline →

Source documents
Source	Document	Type	Tier	Last observed	Snapshots
Hugging Face Hub (public pages, model cards, papers)	huggingface.co/amd/Llama-3.3-70B-Instruct-FP8-KV/raw/main/README.md	model_card	T2· Quality secondary	just now	1
Hugging Face Hub (public pages, model cards, papers)	huggingface.co/amd/Llama-3.3-70B-Instruct-FP8-KV	model_page	T2· Quality secondary	54 min ago	2
Hugging Face Hub (public pages, model cards, papers)	huggingface.co/models?author=amd&p=0&sort=downloads	listing	T2· Quality secondary	1 h ago	2

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