Model HQ

Documentation

Intel Supported Models

Comprehensive list of 225+ AI models optimized for Intel processors using OpenVINO runtime

These models are optimized for Intel processors and use the OpenVINO (OV) runtime for enhanced performance on Intel hardware including CPUs, GPUs, and NPUs.

Intel Optimization Features

Key benefits of Intel-optimized models

Performance Benefits

• Optimized for Intel CPU, GPU, and NPU architectures
• Enhanced inference speed with OpenVINO runtime
• Reduced memory footprint and power consumption
• Hardware-specific optimizations for latest Intel chips

Supported Hardware

• Intel Core processors (Meteor Lake, Lunar Lake, Arrow Lake)
• Intel Xeon processors
• Intel Arc GPUs
• Intel Neural Processing Units (NPUs)

Intel Supported Models

Complete catalog of 228 models optimized for Intel processors

Model Type	Available Models	Parameters
Embedding Models	`industry-bert-contracts-ov`	NA
	`industry-bert-insurance-ov`	NA
	`industry-bert-asset-management-ov`	NA
	`industry-bert-sec-ov`	NA
	`industry-bert-loans-ov`	NA
	`all-mini-lm-l6-v2-ov`	NA
	`all-mpnet-base-v2-ov`	NA
	`paraphrase-multilingual-MiniLM-L12-v2-ov`	NA
	`gte-small-ov`	NA
	`gte-base-ov`	NA
	`gte-large-ov`	NA
	`bge-small-en-v1.5-ov`	NA
	`bge-base-en-v1.5-ov`	NA
	`bge-large-en-v1.5-ov`	NA
	`protectai-prompt-injection-ov`	NA
	`malicious-url-detector-ov`	NA
	`xlm-roberta-language-detector-ov`	NA
	`valurank-bias-ov`	NA
	`unitary-toxic-roberta-ov`	NA
	`jina-reranker-v1-tiny-en-ov`	NA
	`jina-reranker-v1-turbo-en-ov`	NA
	`jina-reranker-tiny-onnx`	NA
	`jina-reranker-turbo-onnx`	NA
	`protectai-prompt-injection-onnx`	NA
	`valurank-bias-onnx`	NA
	`unitary-toxic-roberta-onnx`	NA
Qwen Models	`qwen2-vl-2b-instruct-ov`	2
	`qwen2-vl-7b-instruct-ov`	7
	`bling-qwen-500m-ov`	0.5
	`bling-qwen-1.5b-ov`	1.5
	`dragon-qwen-7b-ov`	7
	`slim-extract-qwen-0.5b-ov`	0.5
	`slim-extract-qwen-1.5b-ov`	1.5
	`qwen2-7b-instruct-ov`	7
	`qwen2-1.5b-instruct-ov`	1.5
	`qwen2-0.5b-chat-ov`	0.5
	`qwen2.5-1.5b-instruct-ov`	1.5
	`qwen3-8b-ov`	8
	`qwen3-1.7b-ov`	1.7
	`qwen3-4b-ov`	4
	`qwen3-14b-ov`	14
	`qwen2.5-0.5b-instruct-ov`	0.5
	`qwen2.5-3b-instruct-ov`	3
	`qwen2.5-14b-instruct-ov`	14
	`qwen2.5-32b-instruct-ov`	32
	`qwen2.5-72b-instruct-ov`	72
	`qwen2.5-coder-7b-instruct-ov`	7
	`bling-qwen-mini-tool`	1.5
	`bling-qwen-0.5b-gguf`	0.5
	`dragon-qwen-7b-gguf`	7
	`qwen2-7B-instruct-gguf`	7
	`qwen3-1.7b-gguf`	1.7
	`qwen3-4b-instruct-gguf`	4
	`qwen3-8b-gguf`	8
	`qwen3-14b-gguf`	14
	`qwen2-1.5b-instruct-gguf`	1.5
	`qwen2-0.5b-instruct-gguf`	0.5
	`slim-extract-qwen-1.5b-gguf`	1.5
	`slim-extract-qwen-nano-gguf`	0.5
	`qwen-2.5-7b-coder-gguf`	7
	`qwen-2.5-14b-instruct-gguf`	14
	`deepseek-qwen-14b-gguf`	14
	`deepseek-qwen-7b-gguf`	7
	`qwen2.5-32b-gguf`	32
Llama-Based Models	`llama-11b-vision-instruct-ov`	11
	`bling-tiny-llama-ov`	1.1
	`bling-tiny-llama-npu-ov`	1.1
	`dragon-llama2-ov`	7
	`llama-2-chat-ov`	7
	`llama-2-13b-chat-ov`	13
	`tiny-llama-chat-ov`	1.1
	`llama-3.1-instruct-ov`	8
	`llama-3.1-8b-instruct-npu-ov`	8
	`nvidia-llama3-chatqa-1.5-8b-ov`	8
	`dolphin-2.9.4-llama3.1-8b-ov`	8
	`llama-3.2-3b-instruct-ov`	3
	`llama-3.2-3b-instruct-npu-ov`	3
	`llama-3.2-1b-instruct-ov`	1.1
	`llama-3.2-1b-instruct-npu-ov`	1.1
	`bling-tiny-llama-onnx`	1.1
	`llama-3.2-3b-onnx-qnn`	3
	`llama-2-chat-onnx`	7
	`llama-3.1-instruct-onnx`	8
	`llama-3.2-1b-instruct-onnx`	1.3
	`llama-3.2-3b-instruct-onnx`	3
	`dragon-llama-3.1-gguf`	8
	`dragon-llama-answer-tool`	7
	`llama-3.1-instruct-gguf`	8
	`llama-2-7b-chat-gguf`	7
	`llama-3-8b-instruct-gguf`	8
	`tiny-llama-chat-gguf`	1.1
	`llama-3.2-1b-instruct-gguf`	1.3
	`llama-3.2-3b-instruct-gguf`	3
Phi Models	`bling-phi-3-ov`	3.8
	`slim-xsum-phi-3-ov`	3.8
	`slim-boolean-phi-3-ov`	3.8
	`slim-sa-ner-phi-3-ov`	3.8
	`slim-summary-phi-3-ov`	3.8
	`slim-sql-phi-3-ov`	3.8
	`slim-extract-phi-3-ov`	3.8
	`phi-3-ov`	3.8
	`phi-3-npu-ov`	3.8
	`phi-4-ov`	14
	`phi-4-mini-ov`	4
	`phi-4-mini-npu-ov`	4
	`phi-4-npu-ov`	14
	`bling-phi-3-onnx`	3.8
	`phi-3-onnx`	3.8
	`phi-3.5-onnx-qnn`	NA
	`phi-3-vision-onnx`	3.8
	`slim-summary-phi-3-onnx`	3.8
	`slim-extract-phi-3-onnx`	3.8
	`slim-boolean-phi-3-onnx`	3.8
	`bling-phi-3-gguf`	3.8
	`bling-phi-3.5-gguf`	3.8
	`phi-3.5-gguf`	3.8
	`phi-4-gguf`	14
	`phi-4-mini-gguf`	4
	`phi-4-mini-reasoning-gguf`	4
	`phi-3-gguf`	3.8
	`slim-extract-phi-3-gguf`	3.8
	`slim-xsum-phi-3-gguf`	3.8
	`slim-boolean-phi-3-gguf`	3.8
	`slim-sa-ner-phi-3-gguf`	3.8
	`slim-q-gen-phi-3-tool`	3.8
	`slim-qa-gen-phi-3-tool`	3.8
Mistral Models	`dragon-mistral-ov`	7.3
	`dragon-mistral-0.3-ov`	7.3
	`mistral-7b-instruct-v0.3-ov`	7.3
	`mistral-7b-v0.3-npu-ov`	7.3
	`mistral-small-instruct-2409-ov`	22
	`mistral-nemo-instruct-2407-ov`	12
	`mistral-7b-instruct-v0.2-ov`	7.3
	`zephyr-mistral-7b-chat-ov`	7.3
	`teknium-open-hermes-2.5-mistral-ov`	7.3
	`dolphin-2.9.3-mistral-7b-32k-ov`	7.3
	`dragon-mistral-0.3-onnx`	7.3
	`mistral-7b-instruct-v0.3-onnx`	7.3
	`dragon-mistral-0.3-gguf`	7.3
	`mistral-3.2-24b-gguf`	24
	`openhermes-2.5-mistral-7b-gguf`	7.3
	`zephyr-7b-beta-gguf`	7.3
	`starling-lm-7b-alpha-gguf`	7
	`dragon-mistral-answer-tool`	7.3
	`mistral-7b-instruct-v0.3-gguf`	7.3
Yi Models	`dragon-yi-6b-ov`	5.8
	`dragon-yi-9b-ov`	8.8
	`yi-9b-chat-ov`	8.8
	`yi-9b-npu-ov`	8.8
	`yi-6b-1.5v-chat-ov`	5.8
	`dragon-yi-9b-gguf`	8.8
	`dragon-yi-answer-tool`	5.8
DRAGON Models	`dragon-qwen-7b-ov`	7
	`dragon-llama2-ov`	7
	`dragon-mistral-ov`	7.3
	`dragon-mistral-0.3-ov`	7.3
	`dragon-yi-6b-ov`	5.8
	`dragon-yi-9b-ov`	8.8
	`dragon-mistral-0.3-onnx`	7.3
	`dragon-llama-3.1-gguf`	8
	`dragon-mistral-0.3-gguf`	7.3
	`dragon-yi-9b-gguf`	8.8
	`dragon-qwen-7b-gguf`	7
	`dragon-yi-answer-tool`	5.8
	`dragon-llama-answer-tool`	7
	`dragon-mistral-answer-tool`	7.3
Slim Models	`slim-sentiment-ov`	1.1
	`slim-sentiment-npu-ov`	1.1
	`slim-xsum-phi-3-ov`	3.8
	`slim-boolean-phi-3-ov`	3.8
	`slim-sa-ner-phi-3-ov`	3.8
	`slim-summary-phi-3-ov`	3.8
	`slim-sql-qwen-base-ov`	2
	`slim-sql-phi-3-ov`	3.8
	`slim-extract-phi-3-ov`	3.8
	`slim-extract-tiny-ov`	1.1
	`slim-extract-tiny-npu-ov`	1.1
	`slim-extract-qwen-0.5b-ov`	0.5
	`slim-extract-qwen-1.5b-ov`	1.5
	`slim-summary-tiny-ov`	1.1
	`slim-summary-tiny-npu-ov`	1.1
	`slim-sql-ov`	1.1
	`slim-sql-npu-ov`	1.1
	`slim-emotions-ov`	1.1
	`slim-emotions-npu-ov`	1.1
	`slim-topics-ov`	1.1
	`slim-topics-npu-ov`	1.1
	`slim-ner-ov`	1.1
	`slim-ner-npu-ov`	1.1
	`slim-intent-ov`	1.1
	`slim-category-ov`	1.1
	`slim-intent-npu-ov`	1.1
	`slim-tags-ov`	1.1
	`slim-tags-npu-ov`	1.1
	`slim-ratings-ov`	1.1
	`slim-ratings-npu-ov`	1.1
	`slim-q-gen-tiny-ov`	1.1
	`slim-qa-gen-tiny-ov`	1.1
	`slim-sentiment-onnx`	1.1
	`slim-extract-tiny-onnx`	1.1
	`slim-summary-tiny-onnx`	1.1
	`slim-sql-onnx`	1.1
	`slim-emotions-onnx`	1.1
	`slim-topics-onnx`	1.1
	`slim-ner-onnx`	1.1
	`slim-intent-onnx`	1.1
	`slim-tags-onnx`	1.1
	`slim-ratings-onnx`	1.1
	`slim-summary-phi-3-onnx`	3.8
	`slim-extract-phi-3-onnx`	3.8
	`slim-boolean-phi-3-onnx`	3.8
	`slim-ner-tool`	1.1
	`slim-sentiment-tool`	1.1
	`slim-emotions-tool`	1.1
	`slim-ratings-tool`	1.1
	`slim-intent-tool`	1.1
	`slim-nli-tool`	1.1
	`slim-topics-tool`	1.1
	`slim-tags-tool`	1.1
	`slim-sql-tool`	1.1
	`bling-answer-tool`	1.1
	`slim-category-tool`	1.1
	`slim-xsum-tool`	1.1
	`slim-extract-tool`	2.8
	`slim-extract-phi-3-gguf`	3.8
	`slim-extract-qwen-1.5b-gguf`	1.5
	`slim-extract-qwen-nano-gguf`	0.5
	`slim-extract-tiny-tool`	1.1
	`slim-summary-tiny-tool`	1.1
	`slim-summary-phi-3-gguf`	1.1
`slim-xsum-phi-3-gguf`	3.8
`slim-boolean-tool`	2.8
`slim-boolean-phi-3-gguf`	3.8
`slim-sa-ner-phi-3-gguf`	3.8
`slim-sa-ner-tool`	2.8
`slim-tags-3b-tool`	2.8
`slim-summary-tool`	2.8
`slim-q-gen-phi-3-tool`	3.8
`slim-q-gen-tiny-tool`	1.1
`slim-qa-gen-tiny-tool`	1.1
`slim-qa-gen-phi-3-tool`	3.8
StableLM Models	`stablelm-zephyr-3b-ov`	2.8
	`stablelm-2-zephyr-1_6b-ov`	1.6
	`stablelm-2-12b-chat-ov`	12
	`bling-stablelm-3b-gguf`	2.8
Gemma Models	`gemma-7b-it-ov`	7
	`codegemma-7b-it-ov`	7
	`gemma-2b-it-ov`	2
	`gemma-2b-it-onnx`	2
	`gemma-3-4b-gguf`	4
	`gemma-3-12b-gguf`	12
	`gemma-2-9b-instruct-gguf`	9
	`gemma-2-27b-instruct-gguf`	27
Specialized Models	`intel-neural-chat-7b-v3-2-ov`	7
	`tiny-dolphin-2.8-1.1b-ov`	1.1
	`dreamgen-wizardlm-2-7b-ov`	7
	`openchat-3.6-8b-20240522-ov`	8
	`mathstral-7b-ov`	7.3
	`whisper-cpp-base-english`	NA
Multimodal Models	`speech-t5-tts-ov`	NA
Multimodal Models	`lcm-dreamshaper-ov`	NA

Getting Started with Intel Models

To use Intel-optimized models in Model HQ:

Ensure you have an Intel processor with OpenVINO support
Select models with the "-ov" suffix from the Models section
The system will automatically use Intel optimizations when available
Monitor performance improvements in the system metrics

Technical Support

For Intel-specific optimization questions or issues, contact our technical support team at support@aibloks.com

🚀 Performance Tip

For optimal performance on Intel hardware, choose models with the highest parameter count that your system can support. Intel's optimization ensures efficient resource utilization.

Check System Requirements