LLM Reference

Gradient Llama 3 Models by Gradient

GradientLlama 3 CommunityOpen weights
4 models2024Up to 1.05m ctxFrom $0.37/1M input

Last refreshed 2026-06-15. Next refresh: weekly.

Details

ResearcherGradient
Commercial useCommercial use: conditional
Models4
Released2024
Max context1.05m

About

Long context windows for Llama 3

Decision facts

Best fit
General model comparison
Capability starting point
Llama 3 8B Gradient 1048K with 1.05m context
Lowest tracked input
Llama 3 8B Gradient 262K · $0.37/1M · Microsoft Foundry
Closest related family
Claude 3

Current Variants

Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.

4 in view

Use when the workload needs 1.05m context and 8B parameters.

2024-041.05m context8B parameters

Use when the workload needs 1.05m context and 70B parameters.

2024-041.05m context70B parameters

Use when the workload needs 1.05m context and 8B parameters.

2024-041.05m context8B parameters

Use when the workload needs 262k context and 8B parameters.

2024-04262k context8B parameters

Release Timeline

1 release group
2024-04
4 current
Llama 3 70B Gradient 1048K
1.05m context70B parameters
Current
Llama 3 8B Gradient 1048K
1.05m context8B parameters
Current
Llama 3 8B Gradient 262K
262k context8B parameters
Current
Llama 3.1 8B Gradient 1048K
1.05m context8B parameters
Current

Specifications(4 models)

Gradient Llama 3 model specifications comparison
ModelReleasedContextParameters
Llama 3 8B Gradient 1048K2024-041.05m8B
Llama 3 70B Gradient 1048K2024-041.05m70B
Llama 3.1 8B Gradient 1048K2024-041.05m8B
Llama 3 8B Gradient 262K2024-04262k8B

Available From(1 provider)

Pricing

Gradient Llama 3 model pricing by provider
ModelProviderInput / 1MOutput / 1MType
Llama 3 8B Gradient 262KMicrosoft Foundry$0.37$1.1Provisioned