Supported self-hosted models and hardware requirements

  • Tier: Ultimate with GitLab Duo Enterprise - Start a trial
  • Offering: Self-managed
  • Status: Beta
History

The following table shows the supported models along with their specific features and hardware requirements to help you select the model that best fits your infrastructure needs for optimal performance.

Approved LLMs

Install one of the following GitLab-approved LLM models:

Model family Model Code completion Code generation GitLab Duo Chat
Mistral Codestral Codestral 22B check-circle Yes check-circle Yes dotted-circle No
Mistral Mistral 7B check-circle Yes dotted-circle No dotted-circle No
Mistral Mistral 7B-it check-circle Yes check-circle Yes check-circle Yes
Mistral Mixtral 8x7B-it check-circle Yes check-circle Yes check-circle Yes
Mistral Mixtral 8x22B-it check-circle Yes check-circle Yes check-circle Yes
Claude 3 Claude 3.5 Sonnet check-circle Yes check-circle Yes check-circle Yes
GPT GPT-3.5-Turbo check-circle Yes check-circle Yes dotted-circle No
GPT GPT-4 Turbo check-circle Yes check-circle Yes dotted-circle No
GPT GPT-4o check-circle Yes check-circle Yes dotted-circle No
GPT GPT-4o-mini check-circle Yes check-circle Yes dotted-circle No

The following models are under evaluation, and support is limited:

Model family Model Code completion Code generation GitLab Duo Chat
CodeGemma CodeGemma 2b check-circle Yes dotted-circle No dotted-circle No
CodeGemma CodeGemma 7b-it dotted-circle No check-circle Yes dotted-circle No
CodeGemma CodeGemma 7b-code check-circle Yes dotted-circle No dotted-circle No
Code Llama Code-Llama 13b-code check-circle Yes dotted-circle No dotted-circle No
Code Llama Code-Llama 13b dotted-circle No check-circle Yes dotted-circle No
DeepSeek Coder DeepSeek Coder 33b Instruct check-circle Yes check-circle Yes dotted-circle No
DeepSeek Coder DeepSeek Coder 33b Base check-circle Yes dotted-circle No dotted-circle No

Hardware Requirements

For optimal performance, the following hardware specifications are recommended as baselines for hosting these models. Hosting requirements may vary depending model to model, so we recommend checking model vendor documentation as well:

  • CPU: Minimum 8 cores (16 threads recommended).
  • RAM: At least 32 GB (64 GB or more recommended for larger models).
  • GPU:
    • Minimum: 2x NVIDIA A100 or equivalent for optimal inference performance.
    • Note: For running Mixtral 8x22B and Mixtral 8x22B-it, it is recommended to use 8x NVIDIA A100 GPUs.
  • Storage: SSD with sufficient space for model weights and data.