Available Models
Install Dell Pro AI Studio Client
Clip-vit-base-patch32arm64EmbeddingNPU - Qualcomm
The CLIP model was developed by researchers at OpenAI to learn about what contributes to robustness in computer vision tasks. The model was also developed to test the ability of models to generalize to arbitrary image classification tasks in a zero-shot manner. The model uses a ViT-B/32 Transformer architecture as an image encoder and uses a masked self-attention Transformer as a text encoder. These encoders are trained to maximize the similarity of (image, text) pairs via a contrastive loss.
- 151M params
- MIT license
- 9 Platforms
Clip-vit-base-patch32x64EmbeddingNPU - Intel
The CLIP model was developed by researchers at OpenAI to learn about what contributes to robustness in computer vision tasks. The model was also developed to test the ability of models to generalize to arbitrary image classification tasks in a zero-shot manner. The model uses a ViT-B/32 Transformer architecture as an image encoder and uses a masked self-attention Transformer as a text encoder. These encoders are trained to maximize the similarity of (image, text) pairs via a contrastive loss.
- 151M params
- MIT license
- 9 Platforms
Devstral-Small-2507x64LLMGPU - Intel
Devstral is an open-source agentic LLM for software engineering. It excels at using tools to explore codebases, editing multiple files and power software engineering agents. NOTE: must install CUDA toolkit v12 to use GPU inference.
- 23.6B params
- Apache 2.0
- 9 Platforms
Granite-4.0-h-smallx64LLMGPU - Nvidia
Granite Small is a 32B parameter instruction-tuned model designed for enterprise-grade AI assistants. It supports multilingual reasoning, summarization, RAG, code generation, and tool-calling with long-context capabilities. NOTE: must install CUDA toolkit v12 to use GPU inference.
- 32.2B params
- Apache 2.0
- 9 Platforms
Granite-4.0-h-tinyx64LLMGPU - Nvidia
Granite Tiny is a 7B parameter MoE instruction-tuned model for multilingual reasoning, summarization, RAG, code, and long-context tasks, optimized for efficient AI assistants and business applications. NOTE: must install CUDA toolkit v12 to use GPU inference.
- 6.67B params
- Apache 2.0
- 9 Platforms
Nomic-embed-text-v1.5arm64EmbeddingCPU
Nomic-embed-text is a high-performing, open-source text embedding model. It supports a maximum context window of 8,192 tokens, making it well-suited for documents, RAG setups, classification, clustering, and more.
- 137M params
- Apache 2.0
- 9 Platforms
Nomic-embed-text-v1.5x64EmbeddingCPU
Nomic-embed-text is a high-performing, open-source text embedding model. It supports a maximum context window of 8,192 tokens, making it well-suited for documents, RAG setups, classification, clustering, and more.
- 137M params
- Apache 2.0
- 9 Platforms
Phi-3.5-mini-instructarm64LLMNPU - Qualcomm
Phi-3.5-mini is a lightweight, state-of-the-art open model built upon datasets used for Phi-3 - synthetic data and filtered publicly available websites - with a focus on very high-quality, reasoning dense data. The model belongs to the Phi-3 model family and supports 128K token context length. The model underwent a rigorous enhancement process, incorporating both supervised fine-tuning, proximal policy optimization, and direct preference optimization to ensure precise instruction adherence and robust safety measures.
- 3.8B params
- MIT license
- 9 Platforms
Phi-3.5-mini-instructx64LLMNPU - AMD
Phi-3.5-mini is a lightweight, state-of-the-art open model built upon datasets used for Phi-3 - synthetic data and filtered publicly available websites - with a focus on very high-quality, reasoning dense data. The model belongs to the Phi-3 model family and supports 128K token context length. The model underwent a rigorous enhancement process, incorporating both supervised fine-tuning, proximal policy optimization, and direct preference optimization to ensure precise instruction adherence and robust safety measures.
- 3.8B params
- MIT license
- 9 Platforms
Phi-3.5-mini-instructx64LLMNPU - Intel
Phi-3.5-mini is a lightweight, state-of-the-art open model built upon datasets used for Phi-3 - synthetic data and filtered publicly available websites - with a focus on very high-quality, reasoning dense data. The model belongs to the Phi-3 model family and supports 128K token context length. The model underwent a rigorous enhancement process, incorporating both supervised fine-tuning, proximal policy optimization, and direct preference optimization to ensure precise instruction adherence and robust safety measures.
- 3.8B params
- MIT license
- 9 Platforms
Whisper-small.enarm64TranscriptionNPU - Qualcomm
Whisper is a pre-trained model for automatic speech recognition (ASR) and speech translation. Trained on 680k hours of labelled data, Whisper models demonstrate a strong ability to generalise to many datasets and domains without the need for fine-tuning. Whisper is a Transformer based encoder-decoder model, also referred to as a sequence-to-sequence model.
- 242M params
- Apache 2.0
- 9 Platforms
Whisper-small.enx64TranscriptionNPU - Intel
Whisper is a pre-trained model for automatic speech recognition (ASR) and speech translation. Trained on 680k hours of labelled data, Whisper models demonstrate a strong ability to generalise to many datasets and domains without the need for fine-tuning. Whisper is a Transformer based encoder-decoder model, also referred to as a sequence-to-sequence model.
- 242M params
- Apache 2.0
- 9 Platforms