DeciLM Models by Deci AI
Last refreshed 2026-05-19. Next refresh: weekly.
Details
About
The DeciLM family of large language models, developed by Deci, specializes in AI-driven innovations for enhancing AI capabilities. These models are particularly renowned for their efficiency and accuracy, especially in the 6 and 7 billion parameter range 1512. The DeciLM-7B model is a prime example, excelling in speed and accuracy, even outperforming some models with greater parameters 25. Utilizing Deci’s AutoNAC technology and Grouped-Query Attention, these models are optimized for precision and computational efficiency 18. The availability of an "instruct" iteration, DeciLM-7B-instruct, further enhances instruction handling 3. Under the Apache 2.0 license, the DeciLM models are accessible for diverse applications, benefiting from the speed enhancements offered by the Infery-LLM inference engine 2.
Decision facts
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
Use when the workload needs 8k context and 7B parameters.
Use when the workload needs 4k context and 6B parameters.
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| DeciLM 7B | Use when the workload needs 8k context and 7B parameters. | 2024-01 | 8k context7B parameters | Current |
| DeciLM 6B | Use when the workload needs 4k context and 6B parameters. | 2024-01 | 4k context6B parameters | Current |
Release Timeline
1 release groupSpecifications(2 models)
Available From(1 provider)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| DeciLM 7B | Microsoft Foundry | $0.52 | $0.67 | Provisioned |
