@@ -50,18 +50,18 @@ Support for the following GitLab-supported large language models (LLMs) is gener
50
50
| GPT | [ GPT-4 Turbo] ( https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models?tabs=python-secure#gpt-4 ) | [ Azure OpenAI] ( https://learn.microsoft.com/en-us/azure/ai-services/openai/overview ) | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="check-circle-dashed" >}} Largely compatible |
51
51
| GPT | [ GPT-4o] ( https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models?tabs=python-secure#gpt-4o-and-gpt-4-turbo ) | [ Azure OpenAI] ( https://learn.microsoft.com/en-us/azure/ai-services/openai/overview ) | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="check-circle-filled" >}} Fully compatible |
52
52
| GPT | [ GPT-4o-mini] ( https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/models?tabs=python-secure#gpt-4o-and-gpt-4-turbo ) | [ Azure OpenAI] ( https://learn.microsoft.com/en-us/azure/ai-services/openai/overview ) | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="check-circle-dashed" >}} Largely compatible |
53
+ | Llama | [ Llama 3 8B] ( https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | {{< icon name="check-circle-dashed" >}} Largely compatible | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="dash-circle" >}} Not compatible |
54
+ | Llama | [ Llama 3.1 8B] ( https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | {{< icon name="check-circle-dashed" >}} Largely compatible | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="check-circle-dashed" >}} Largely compatible |
55
+ | Llama | [ Llama 3 70B] ( https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | {{< icon name="check-circle-dashed" >}} Largely compatible | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="dash-circle" >}} Not compatible |
56
+ | Llama | [ Llama 3.1 70B] ( https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | {{< icon name="check-circle-dashed" >}} Largely compatible | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="check-circle-filled" >}} Fully compatible |
57
+ | Llama | [ Llama 3.3 70B] ( https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | {{< icon name="check-circle-dashed" >}} Largely compatible | {{< icon name="check-circle-filled" >}} Fully compatible | {{< icon name="check-circle-filled" >}} Fully compatible |
53
58
54
59
### Experimental and beta models
55
60
56
61
The following models are configurable for the functionalities marked below, but are currently in beta or experimental status, under evaluation, and are excluded from the "Customer Integrated Models" definition in the [ AI Functionality Terms] ( https://handbook.gitlab.com/handbook/legal/ai-functionality-terms/ ) :
57
62
58
63
| Model family | Model | Supported platforms | Status | Code completion | Code generation | GitLab Duo Chat |
59
64
| --------------- | -------| ---------------------| --------| -----------------| -----------------| -----------------|
60
- | Llama | [ Llama 3 8B] ( https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | Beta | {{< icon name="check-circle" >}} Yes | {{< icon name="check-circle" >}} Yes | {{< icon name="check-circle" >}} Yes |
61
- | Llama | [ Llama 3.1 8B] ( https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | Beta | {{< icon name="check-circle" >}} Yes | {{< icon name="check-circle" >}} Yes | {{< icon name="check-circle" >}} Yes |
62
- | Llama | [ Llama 3 70B] ( https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | Beta | {{< icon name="check-circle" >}} Yes | {{< icon name="check-circle" >}} Yes | {{< icon name="check-circle" >}} Yes |
63
- | Llama | [ Llama 3.1 70B] ( https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | Beta | {{< icon name="check-circle" >}} Yes | {{< icon name="check-circle" >}} Yes | {{< icon name="check-circle" >}} Yes |
64
- | Llama | [ Llama 3.3 70B] ( https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | Beta | {{< icon name="check-circle" >}} Yes | {{< icon name="check-circle" >}} Yes | {{< icon name="check-circle" >}} Yes |
65
65
| CodeGemma | [ CodeGemma 2b] ( https://huggingface.co/google/codegemma-2b ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | Experimental | {{< icon name="check-circle" >}} Yes | {{< icon name="dotted-circle" >}} No | {{< icon name="dotted-circle" >}} No |
66
66
| CodeGemma | [ CodeGemma 7b-it] ( https://huggingface.co/google/codegemma-7b-it ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | Experimental | {{< icon name="dotted-circle" >}} No | {{< icon name="check-circle" >}} Yes | {{< icon name="dotted-circle" >}} No |
67
67
| CodeGemma | [ CodeGemma 7b-code] ( https://huggingface.co/google/codegemma-7b ) | [ vLLM] ( supported_llm_serving_platforms.md#for-self-hosted-model-deployments ) | Experimental | {{< icon name="check-circle" >}} Yes | {{< icon name="dotted-circle" >}} No | {{< icon name="dotted-circle" >}} No |
0 commit comments