Skip to content
AI Atlas
ModelActive

amazon/GPT-OSS-20B-P-EAGLE

Amazon Web Serviceshuggingface.co/amazon/GPT-OSS-20B-P-EAGLE

Open in Graph
data quality53

Updated 12 h ago · first seen 11 Sept 2026

model_01M294X4D6DH08BQVXZ8VRZWPF

Overview

Identity

Identity block not returned by the API for this entity.

Openness

Openness not classified yet — no sourced evidence to place this model in the ontology.

Key facts

Release date

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

Architecture

Architecture

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

Model type

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

Parameters

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

Weights dtype

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

File size

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

Hugging Face repo

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

Hardware fit37

Estimated

37 of 37 device × quantization combinations fit.

Run locally: your machine →
Estimated hardware fit
HardwareQuantizationDevice memoryEst. memoryFits
Apple M3 Ultra4bit1.5 GB est.Yes
Mac Studio (Apple M5 Ultra)4bit1.5 GB est.Yes
Apple M2 Ultra4bit1.5 GB est.Yes
Apple M1 Ultra4bit1.5 GB est.Yes
Apple M3 Max4bit1.5 GB est.Yes
Apple M4 Max4bit1.5 GB est.Yes
Mac Studio (Apple M5 Max)4bit1.5 GB est.Yes
MacBook Pro (Apple M5 Max)4bit1.5 GB est.Yes
Apple M2 Max4bit1.5 GB est.Yes
Apple M1 Max4bit1.5 GB est.Yes
Apple M4 Pro4bit1.5 GB est.Yes
Mac mini (Apple M5 Pro)4bit1.5 GB est.Yes
MacBook Pro (Apple M5 Pro)4bit1.5 GB est.Yes
Apple M3 Pro4bit1.5 GB est.Yes
Apple M1 Pro4bit1.5 GB est.Yes
Apple M2 Pro4bit1.5 GB est.Yes
Apple M44bit1.5 GB est.Yes
iMac (Apple M4)4bit1.5 GB est.Yes
Mac mini (Apple M6)4bit1.5 GB est.Yes
MacBook Air (Apple M5)4bit1.5 GB est.Yes
MacBook Pro (Apple M5)4bit1.5 GB est.Yes
Apple M24bit1.5 GB est.Yes
Apple M34bit1.5 GB est.Yes
Apple M14bit1.5 GB est.Yes
NVIDIA GeForce RTX 30904bit24 GB1.5 GB est.Yes
NVIDIA GeForce RTX 40904bit24 GB1.5 GB est.Yes
NVIDIA GeForce RTX 50904bit32 GB1.5 GB est.Yes
NVIDIA A100 80GB4bit80 GB1.5 GB est.Yes
NVIDIA H100 SXM4bit80 GB1.5 GB est.Yes
NVIDIA H100 NVL4bit94 GB1.5 GB est.Yes
NVIDIA DGX Spark4bit128 GB1.5 GB est.Yes
NVIDIA H2004bit141 GB1.5 GB est.Yes
NVIDIA H200 NVL4bit141 GB1.5 GB est.Yes
NVIDIA B2004bit180 GB1.5 GB est.Yes
AMD Instinct MI300X4bit192 GB1.5 GB est.Yes
AMD Instinct MI325X4bit256 GB1.5 GB est.Yes
NVIDIA DGX B2004bit1,440 GB1.5 GB est.Yes
Assumptions (6)
  • Estimated, not measured: weights = parameters × bytes/param × 1.15 runtime overhead.
  • bytes/param: 4bit = 0.5, 8bit = 1.0, fp16 = 2.0 (uniform quantization, no per-layer exceptions).
  • KV cache approximated at 0.5 GB per 8 192 tokens of context, independent of architecture (GQA/MLA models need less).
  • A model 'fits' when the estimate is at most the device memory minus 2 GB reserved for the OS and framework.
  • Mixture-of-experts models are estimated on total parameters (all experts must be resident); active parameters are ignored.
  • Device memory uses the largest configuration when several are listed (e.g. Apple silicon tiers).

Papers1

Change history16

Temporal, append-only claims: a new observation closes the previous claim instead of overwriting it. Rewind the record with the as-of picker.

Viewing AI Atlas as of 12 Sept 2026 — attributes exactly as the atlas knew them on that day; later corrections are not shown.

Back to today →

Attributes as of 12 Sept 2026 20 claims in force

Release date
20 Feb 2026
Openness
open-weights
License
Apache-2.0
Architecture
LlamaForCausalLM
Parameters
1.8B
File size
3.6 GB
Hugging Face repo
amazon/GPT-OSS-20B-P-EAGLE
Downloads
368
Likes
1
Commercial use allowed
Yes
Derivatives allowed
Yes
Gated
No
Last modified
2026-05-01T22:07:24+00:00
Downloads all time
4,793
Model type
llama
Redistribution allowed
Yes
Tags
safetensors, llama
Weights available
Yes
Weights dtype
BF16, BOOL, I64
16 claims · 16 properties

Release daterelease_date1

Claim history for Release date
ValueValid from → toStatusSourceConfidenceExtractor
20 Feb 2026currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Opennessopenness1

Claim history for Openness
ValueValid from → toStatusSourceConfidenceExtractor
open-weightscurrentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Licenselicense1

Claim history for License
ValueValid from → toStatusSourceConfidenceExtractor
apache-2.0currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Architecturearchitecture1

Claim history for Architecture
ValueValid from → toStatusSourceConfidenceExtractor
LlamaForCausalLMcurrentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Parametersparameter_count1

Claim history for Parameters
ValueValid from → toStatusSourceConfidenceExtractor
1.8BcurrentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

File sizefile_size_gb1

Claim history for File size
ValueValid from → toStatusSourceConfidenceExtractor
3.6 GBcurrentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Hugging Face repohf_repo1

Claim history for Hugging Face repo
ValueValid from → toStatusSourceConfidenceExtractor
amazon/GPT-OSS-20B-P-EAGLEcurrentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Model cardmodel_card_url1

Claim history for Model card
ValueValid from → toStatusSourceConfidenceExtractor
https://huggingface.co/amazon/GPT-OSS-20B-P-EAGLEcurrentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Downloadsmetric.downloads1

Claim history for Downloads
ValueValid from → toStatusSourceConfidenceExtractor
368currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Likesmetric.likes1

Claim history for Likes
ValueValid from → toStatusSourceConfidenceExtractor
1currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Gatedgated1

Claim history for Gated
ValueValid from → toStatusSourceConfidenceExtractor
NocurrentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Last modifiedlast_modified1

Claim history for Last modified
ValueValid from → toStatusSourceConfidenceExtractor
2026-05-01T22:07:24+00:00currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Downloads all timemetric.downloads_all_time1

Claim history for Downloads all time
ValueValid from → toStatusSourceConfidenceExtractor
4,793currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Model typemodel_type1

Claim history for Model type
ValueValid from → toStatusSourceConfidenceExtractor
llamacurrentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Tagstags1

Claim history for Tags
ValueValid from → toStatusSourceConfidenceExtractor
safetensors, llamacurrentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

Weights dtypeweights_dtype1

Claim history for Weights dtype
ValueValid from → toStatusSourceConfidenceExtractor
BF16, BOOL, I64currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

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 →

Provenance

Attributed facts

16

Source tiers

T216

Freshest observation

12 h ago

Conflicts

None

Source documents 3

Source documents
SourceDocumentTypeTierLast observedSnapshots
Hugging Face Hub (public pages, model cards, papers)huggingface.co/amazon/GPT-OSS-20B-P-EAGLE/raw/main/README.md model_cardT2· Quality secondary11 h ago1
Hugging Face Hub (public pages, model cards, papers)huggingface.co/amazon/GPT-OSS-20B-P-EAGLE model_pageT2· Quality secondary12 h ago2
Hugging Face Hub (public pages, model cards, papers)huggingface.co/models?author=amazon&p=0&sort=downloads listingT2· Quality secondary12 h ago2

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

Data quality (53/100) measures how well AI Atlas knows this entity — completeness, primary-source ratio, freshness, conflicts — never how good the model is. Methodology →