LLM AtlasLLM AtlasSearch models

Provider analysis

Meta

Publisher of the Llama family and related safety models distributed as open-weight releases.

Last verified: 2026-03-29Confidence: HighPrimary sources: 2

This provider page blends full-profile entries with broader verified listings. Use it to separate deeply evaluated flagship models from source-backed records that are tracked primarily for market visibility, access data, and freshness coverage.

Headquarters
Menlo Park, CA
Founded
2004
Models tracked
17
Full-profile models
17
Catalog last verified
2026-03-29
Latest model verification
2026-03-29
Newest release tracked
2025-04-05
Confidence
High
Access mix
open-weight, self-hosted, hosted
API models
0

Tracked models available through provider-managed APIs.

Open-weight models
17

Models with downloadable weights or self-hosted distribution paths.

Primary source links
36

Total source references attached across this provider catalog.

Provider sources

Official links used to verify the provider profile and platform coverage.

Last verified: 2026-03-29

Meta

Llama 4 Maverick

Llama 4

Meta's 17Bx128E MoE open-weight model with 1M token context, pretrained on ~22T tokens. Strong multimodal and multilingual capabilities for teams that need control, private deployment, and customization.

Score 843 sources
textvisionopen-sourcecodeopen-weightself-hostedhosted
Context
1,048,576
Input
$0.0008/1K tok
Output
$0.002/1K tok
Coverage
Full profile

Meta

Llama 4 Scout

Llama

Meta's Llama 4 Scout (17Bx16E MoE, 109B total params) with an extraordinary 10M token context window.

Score 732 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
10,485,760
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Llama 3.3 70B Instruct

Llama

Meta's latest Llama 3.x dense model, 70B parameters with 128K context.

Score 742 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
128,000
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Llama 3.2 3B Instruct

Llama

Meta's lightweight Llama 3.2 models for edge and mobile deployment with 128K context.

Score 612 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
128,000
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Llama 3.2 1B Instruct

Llama

Meta's lightweight Llama 3.2 models for edge and mobile deployment with 128K context.

Score 612 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
128,000
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Llama 3.1 405B Instruct

Llama

Meta's Llama 3.1 405B flagship with 128K context.

Score 702 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
128,000
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Llama 3.1 70B Instruct

Llama

Meta's Llama 3.1 70B with 128K context for instruction-following.

Score 702 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
128,000
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Llama 3.1 8B Instruct

Llama

Meta's Llama 3.1 8B with 128K context for efficient deployment.

Score 642 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
128,000
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Meta Llama 3 70B Instruct

Llama

Meta's original Llama 3 and Code Llama 70B models with 8K context.

Score 602 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
8,192
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Code Llama 70B Instruct

Llama

Meta's original Llama 3 and Code Llama 70B models with 8K context.

Score 602 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
8,192
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Meta Llama 3 8B Instruct

Llama

Meta's smaller Llama 3 and Code Llama models with 8K context.

Score 572 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
8,192
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Code Llama 34B Instruct

Llama

Meta's smaller Llama 3 and Code Llama models with 8K context.

Score 572 sources
textcodeopen-sourceopen-weightself-hostedhosted
Context
8,192
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Llama 3.2 90B Vision Instruct

Llama multimodal

Meta's Llama 3.2 90B vision model with 128K context for image understanding.

Score 693 sources
textvisionopen-sourceopen-weightself-hostedhosted
Context
128,000
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Llama 3.2 11B Vision Instruct

Llama multimodal

Meta's Llama 3.2 11B vision model for efficient multimodal understanding.

Score 642 sources
textvisionopen-sourceopen-weightself-hostedhosted
Context
128,000
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Llama Guard 4 12B

Llama Guard

Meta's Llama Guard 4 (12B) safety model for filtering unsafe prompts.

Score 602 sources
textvisionopen-sourceopen-weightself-hostedhosted
Context
131,072
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Llama Guard 3 11B Vision

Llama Guard

Meta's Llama Guard 3 vision safety model for multimodal content moderation.

Score 572 sources
textvisionopen-sourceopen-weightself-hostedhosted
Context
131,072
Input
Not applicable
Output
Not applicable
Coverage
Full profile

Meta

Prompt Guard 86M

Llama Guard

Meta's lightweight 86M Prompt Guard for detecting jailbreak and prompt injection attempts.

Score 412 sources
textvisionopen-sourceopen-weightself-hostedhosted
Context
512
Input
Not applicable
Output
Not applicable
Coverage
Full profile