Open Model Weights
Home / Models / Meta-Llama-3-8B

META · FIELD-VERIFIED

Meta-Llama-3-8B

Checked against the listed Hugging Face source repository on 2026-10-03. Unknown values are left explicit rather than inferred.

Source repository ↗Compare this model →History & diff →Field-verified 2026-10-03Machine-readable JSON ↗
DeveloperMeta
Parameters8.03026B
ContextNot declared in checked config/model card
Weight files6
Licensellama3
Accessgated (manual)
Evidence statusSource-first verified
Repository revision checked2026-10-03
Full field verification2026-10-02
Source policy →Methodology →

MODEL INTELLIGENCE · SOURCE-FIRST

What is Meta-Llama-3-8B?

OMW evidence review3 Oct 2026

Open Model Weights records Meta-Llama-3-8B as a 8.03026B-parameter text open-weight model in the Meta-Llama-3 family, with Meta listed as the developer. The checked source repository provides 6 recognized weight artifacts, format evidence for Safetensors, PyTorch and a verified context window of Not declared in checked config/model card. Values that are not evidenced in the checked sources remain explicitly undisclosed or unknown.

The repository declares llama3. Commercial use allowed subject to Llama community-license conditions. Open Model Weights reports the checked license evidence and classification; this is not legal advice.

No base model is declared in the checked standard lineage metadata, so Open Model Weights does not infer one. In the checked repository, training or recipe evidence is present and a data-disclosure signal is present.

Source repository ↗Model card ↗Configuration ↗License evidence ↗

DEPLOYMENT PROFILE

Weight memory, without pretending it is a deployment guarantee.

These bars visualize weight-only decimal-GB estimates. They exclude KV cache, activations, runtime overhead, optimizer state and sharding.

WEIGHT-ONLY MEMORY
BF16 / FP16~16.1 GB
FP8 / INT8~8.0 GB
INT4~4.0 GB
FIELD EVIDENCE MAP
WeightsVerified
LicenseVerified
Commercial useClassified
ContextNot Disclosed
FormatsVerified
LineageNot declared
Training recipeTraining described in model card
Training dataDescribed in model card
RuntimeSource-derived · not runtime-tested

LICENSE & USE

llama3

Commercial use allowed subject to Llama community-license conditions

Declared licensellama3
ClassificationAllowed with conditions
VerificationLicense file checked

This section summarizes the checked repository evidence. It is not legal advice. Inspect license evidence ↗

LINEAGE & PROVENANCE

What the checked sources disclose — and what they do not.

Open Model Weights does not infer a hidden base model or training corpus when the checked sources do not declare one.

LINEAGE
BASE / PARENTBase model not declaredUnknown
not inferred
CURRENT RECORDMeta-Llama-3-8BMeta-Llama-3
TRAINING DISCLOSURE
Training recipeTraining described in model card
Training dataDescribed in model card
FamilyMeta-Llama-3

RUNTIME SIGNALS

Compatibility claims stay attached to their evidence level.

Declared and model-card-mentioned runtimes are discovery signals. They are not treated as Open Model Weights runtime tests unless explicitly measured.

DeclaredTransformers
MentionedTGI
OMW runtime testNot performed

EVIDENCE TRAIL

Sources and observed history.

This is where the article remains auditable: source links on one side, the registry's observed change history on the other.

PrimaryModel Repositoryhttps://huggingface.co/meta-llama/Meta-Llama-3-8BInspect source ↗PrimaryModel Apihttps://huggingface.co/api/models/meta-llama/Meta-Llama-3-8BInspect source ↗PrimaryModel Cardhttps://huggingface.co/meta-llama/Meta-Llama-3-8B/blob/main/README.mdInspect source ↗PrimaryConfighttps://huggingface.co/meta-llama/Meta-Llama-3-8B/blob/main/config.jsonInspect source ↗Primary Or DeclaredLicensehttps://www.llama.com/llama3/license/Inspect source ↗
Repository revision check

Listed Hugging Face source repository SHA unchanged; prior field evidence retained.

Repository revision check

Listed Hugging Face source repository SHA unchanged; prior field evidence retained.

Field-by-field verification

Checked Hugging Face API metadata, exact repository file list, model card and config when accessible.

Registry record created

Previous verification level: discovered-pending.

RELATED RECORDS

Continue through structurally related records.

Family matches are preferred; otherwise the fallback requires the same developer, model modality and declared pipeline task. No quality score or recommendation rank is used.

Same developer + taskLlama 3.3 70B Instruct

71B · Not declared in checked config/model card

Same developer + taskLlama-2-7b-chat-hf

6.73842B · Not declared in checked config/model card

Same developer + taskLlama-2-7b-hf

6.73842B · Not declared in checked config/model card

Same developer + taskLlama-3.1-405B-FP8

406B · Not declared in checked config/model card

FIELD-BY-FIELD RECORD

Inspect the underlying evidence.

The editorial layer above summarizes only what the verified registry can support. The sections below expose the exact fields, artifacts, source links and verification states.

WeightsLicenseHardwareFormatsLineageTrainingRuntimeSourcesVerification

01 · WEIGHTS

Exact repository artifacts

Statusverified-available
Accessgated (manual)
Repositorymeta-llama/Meta-Llama-3-8B
Recognized weight files6
Show all 6 exact weight files
HF API file listing ↗

02 · LICENSE

License & commercial use

Declared licensellama3
License IDllama3
Commercial useCommercial use allowed subject to Llama community-license conditions
Repository license filesLICENSE

Classification describes the checked license terms; it is not legal advice.

License / model-card evidence ↗

03 · HARDWARE

Weight-memory estimate

BF16 / FP16~16.1 GB
FP8 / INT8~8.0 GB
INT4~4.0 GB

Weight-only decimal-GB estimate from total parameter count; excludes KV cache, activations, runtime overhead, optimizer state and sharding.

04 · FORMATS

Formats & precision evidence

SafetensorsPyTorch
BF16: observedFP16: not observedFP8: not observedINT8: not observedINT4: not observedGGUF: not observed

“Not observed” means no matching repository artifact/name/dtype signal was found; it does not prove no third-party conversion exists.

05 · LINEAGE

Base model & family

FamilyMeta-Llama-3
VariantMeta-Llama-3-8B
Declared base modelNot declared in model-card metadata
RelationNot declared

06 · TRAINING ASSETS

Recipe & data disclosure

Training recipetraining-described-in-model-card
Data disclosuredescribed-in-model-card
EvaluationsNot classified by automated verifier
Intermediate checkpointsNot classified by automated verifier

07 · RUNTIME SUPPORT

Declared and mentioned runtimes

Declared libraryTransformers
Declared/taggedTransformers
Mentioned in model cardTGI
Artifact signalsNone
Tested by Open Model WeightsNo — source-derived support only

Open Model Weights does not claim runtime compatibility unless it is declared, tagged, mentioned by the publisher, or evidenced by an artifact. No inference benchmark is implied.

08 · SOURCES

Field evidence

Primary / declared sources5
Repository revision8cde5ca83804
Source policyHow evidence is classified →
model_repositoryhttps://huggingface.co/meta-llama/Meta-Llama-3-8B
model_apihttps://huggingface.co/api/models/meta-llama/Meta-Llama-3-8B
model_cardhttps://huggingface.co/meta-llama/Meta-Llama-3-8B/blob/main/README.md
confighttps://huggingface.co/meta-llama/Meta-Llama-3-8B/blob/main/config.json
licensehttps://www.llama.com/llama3/license/

09 · VERIFICATION HISTORY

Checked, not assumed

Repository revision check

listed Hugging Face source repository SHA unchanged; prior field evidence retained.

Repository revision check

listed Hugging Face source repository SHA unchanged; prior field evidence retained.

Field-by-field verification

Checked Hugging Face API metadata, exact repository file list, model card and config when accessible.

Registry record created

Previous verification level: discovered-pending.

licenseverified
commercial useclassified
contextnot-disclosed
weight filesverified
formatsverified
base modelnot-declared
training recipetraining-described-in-model-card
data disclosuredescribed-in-model-card
runtime supportsource-derived-not-runtime-tested