LLM Reference

DeciCoder Models by Deci AI

Deci AIApache 2.0Open sourceCoding
2 models2023Up to 4k ctxFrom $0.07/1M input

Last refreshed 2026-05-19. Next refresh: weekly.

Details

ResearcherDeci AI
LicenseApache 2.0OSI-approved
Commercial useCommercial use: permitted
Models2
Released2023
Max context4k

About

The DeciCoder family consists of open-source large language models (LLMs) focused on code generation, developed by Deci AI. These models leverage Deci's proprietary AutoNAC technology to optimize their architecture for enhanced accuracy and speed 52. The flagship model, DeciCoder-1b, features 1 billion parameters and is specifically designed for code completion tasks. Trained on subsets of the Starcoder Training Dataset, it supports languages like Python, Java, and JavaScript and is equipped with Grouped Query Attention and a 2048-token context window 2. Designed to efficiently generate accurate code, DeciCoder models offer cost savings by running effectively on more affordable hardware, with plans for expanding the model lineup 53.

Decision facts

Best fit
coding
Capability starting point
DeciCoder 6B with 4k context
Lowest tracked input
DeciCoder 1B · $0.07/1M · Microsoft Foundry
Closest related family
SWE-1

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 4k context and 6B parameters.

2023-084k context6B parameters

Use when the workload needs 4k context and 1B parameters.

2023-084k context1B parameters

Release Timeline

1 release group
2023-08
2 current
DeciCoder 1B
4k context1B parameters
Current
DeciCoder 6B
4k context6B parameters
Current

Specifications(2 models)

DeciCoder model specifications comparison
ModelReleasedContextParameters
DeciCoder 6B2023-084k6B
DeciCoder 1B2023-084k1B

Available From(1 provider)

Pricing

DeciCoder model pricing by provider
ModelProviderInput / 1MOutput / 1MType
DeciCoder 1BMicrosoft Foundry$0.07$0.07Provisioned