Euryale Models by Sao10K

Sao10KCC-BY-NC-4.0Open weights
2 models2024Up to 131k ctxFrom $0.850/1M input

Last refreshed 2026-09-30. Next refresh: weekly.

Details

ResearcherSao10K
Commercial useCommercial use: non-commercial
Models2
Released2024
Max context131k

About

Euryale is Sao10K's series of community fine-tunes of Meta Llama models for creative writing and roleplay. The series spans Llama 3 (v2.1) and Llama 3.1 (v2.2) bases, with training focused on multi-turn coherency and narrative quality.

Decision facts

Best fit
General model comparison
Capability starting point
L3.1 70B Euryale V2.2 with 131k context
Lowest tracked input
L3.1 70B Euryale V2.2 · $0.850/1M · OpenRouter
Closest related family
Lunaris

Current Variants

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

2 in view

Use when the workload needs 131k context and 70B parameters.

2024-08131k context70B parameters

Use when the workload needs 8k context and 70B parameters.

2024-068k context70B parameters

Release Timeline

2 release groups
2024-08
1 current
L3.1 70B Euryale V2.2
131k context70B parameters
Current
2024-06
1 current
L3 70B Euryale V2.1
8k context70B parameters
Current

Specifications(2 models)

Euryale model specifications comparison
ModelReleasedContextParameters
L3.1 70B Euryale V2.22024-08131k70B
L3 70B Euryale V2.12024-068k70B

Available From(2 providers)

Pricing

Euryale model pricing by provider
ModelProviderInput / 1MOutput / 1MType
L3.1 70B Euryale V2.2OpenRouter$0.85$0.85Serverless
L3 70B Euryale V2.1Novita AI$1.48$1.48Serverless
L3.1 70B Euryale V2.2Novita AI$1.48$1.48Serverless