Local

1. Docs
2. /Reference
3. /Search C# APIs
4. /Aspire.Hosting.Foundry
5. /Local

Classstaticnet10.0

📦 Aspire.Hosting.Foundry v13.6.0-preview.1.26478.8

Models available on Foundry Local.

namespace Aspire.Hosting.Foundry;

public static class Local
{
    // ...
}

Source

Search members of Local...

44 members

Clear filtersReset all

Fields44

DeepseekR114bstatic

FoundryModel

This model is an optimized version of DeepSeek-R1-Distill-Qwen-14B to enable local inference on Intel GPUs.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is a conversion of the DeepSeek-R1-Distill-Qwen-14B for local inference on Intel GPUs.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model DeepSeek-R1-Distill-Qwen-14B for details.

DeepseekR115bstatic

FoundryModel

This model is an optimized version of DeepSeek-R1-Distill-Qwen-1.5B to enable local inference on Intel GPUs.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is a conversion of the DeepSeek-R1-Distill-Qwen-1.5B for local inference on Intel GPUs.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model DeepSeek-R1-Distill-Qwen-1.5B for details.

DeepseekR17bstatic

FoundryModel

This model is an optimized version of DeepSeek-R1-Distill-Qwen-7B to enable local inference on Intel GPUs.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is a conversion of the DeepSeek-R1-Distill-Qwen-7B for local inference on Intel GPUs.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model DeepSeek-R1-Distill-Qwen-7B for details.

Gemma4E2bItstatic

FoundryModel

This model is an optimized version of Gemma 4 E2B IT for local inference. This optimized model is published here in ONNX format to run on CUDA-capable GPU devices, with the precision best suited to this target.

ONNX Models

Here are some of the optimized configurations we have added:

1. ONNX model for CUDA GPU using RTN quantization.

Model Description

• Developed by: Google DeepMind

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of Gemma 4 E2B IT for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Gemma 4 E2B IT for details.

GptOss20bstatic

FoundryModel

This model is an optimized version of gpt-oss-20b to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: Apache-2.0

• License Description: Use of this model is subject to the terms of the Apache License, Version 2.0, available at http://www.apache.org/licenses/LICENSE-2.0.

• Model Description: This is a conversion of the gpt-oss-20b model for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Azure AI Foundry model gpt-oss-20b for details.

Ministral33bInstruct2512static

FoundryModel

This model is an optimized version of Ministral-3-3B-Instruct-2512 for local inference. This optimized model is published here in ONNX format to run on CUDA-capable GPU devices, with the precision best suited to this target.

ONNX Models

Here are some of the optimized configurations we have added:

1. ONNX model for CUDA GPU using RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of Ministral-3-3B-Instruct-2512 for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Ministral-3-3B-Instruct-2512 for details.

Mistral7bV02static

FoundryModel

This model is an optimized version of Mistral-7B-Instruct-v0.2 to enable local inference on Intel GPUs.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Mistral-7B-Instruct-v0.2 for local inference on Intel GPUs.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Mistral-7B-Instruct-v0.2 for details.

MistralNemo12bInstructstatic

FoundryModel

This model is an optimized version of Mistral-Nemo-Instruct-2407 to enable local inference. This model uses GPTQ quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Mistral-Nemo-Instruct-2407 for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Mistral-Nemo-Instruct-2407 for details.

Nemotron35AsrStreaming06bstatic

FoundryModel

This model is an optimized version of nemotron-3.5-asr-streaming-0.6b to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is an optimized version of the nemotron-3.5-asr-streaming-0.6b model for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model nemotron-3.5-asr-streaming-0.6b for details.

NemotronSpeechStreamingEn06bstatic

FoundryModel

This model is an optimized version of nemotron-speech-streaming-en-0.6b to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is an optimized version of the nemotron-speech-streaming-en-0.6b model for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model nemotron-speech-streaming-en-0.6b for details.

NemotronSpeechStreamingEs06bstatic

FoundryModel

This model is a fine-tuned and optimized derivative of nemotron-speech-streaming-en-0.6b, adapted for Spanish speech recognition. The model is optimized for local inference.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This model is derived from nemotron-speech-streaming-en-0.6b and has been fine-tuned for Spanish speech recognition. It has been converted and optimized for efficient local inference.

• Disclaimer: This model is a fine-tuned and optimized derivative of the base model, adapted for Spanish speech recognition. Due to this adaptation, its capabilities differ substantially from the original English model and are specialized for Spanish-language use cases. The model may not perform as expected on other languages or tasks. Users are responsible for evaluating the model in their specific application context.

Base Model Information

See Hugging Face model https://huggingface.co/nvidia/nemotron-speech-streaming-en-0.6b for details.

Olmo37bInstructstatic

FoundryModel

This model is an optimized version of Olmo-3-7B-Instruct to enable local inference. This model uses GPTQ quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Olmo-3-7B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Olmo-3-7B-Instruct for details.

ParakeetTdt06bV2static

FoundryModel

This model is an optimized version of parakeet-tdt-0.6b-v2 for local inference. This optimized model is published here in ONNX format to run on CUDA-capable GPU devices, with the precision best suited to this target.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is a conversion of the parakeet-tdt-0.6b-v2 for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model parakeet-tdt-0.6b-v2 for details.

Phi35Ministatic

FoundryModel

This model is an optimized version of Phi-3.5-mini-instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is a conversion of the Phi-3.5-mini-instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Phi-3.5-mini-instruct for details.

Phi3Mini128kstatic

FoundryModel

This model is an optimized version of Phi-3-Mini-128K-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is a conversion of the Phi-3-Mini-128K-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Phi-3-Mini-128K-Instruct for details.

Phi3Mini4kstatic

FoundryModel

This model is an optimized version of Phi-3-Mini-4K-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is a conversion of the Phi-3-Mini-4K-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Phi-3-Mini-4K-Instruct for details.

Phi4static

FoundryModel

This model is an optimized version of Phi-4 to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is a conversion of the Phi-4 for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Phi-4 for details.

Phi4Ministatic

FoundryModel

This model is an optimized version of Phi-4-mini-instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is a conversion of the Phi-4-mini-instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Phi-4-mini-instruct for details.

Phi4MiniReasoningstatic

FoundryModel

This model is an optimized version of Phi-4-mini-reasoning to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: MIT

• Model Description: This is a conversion of the Phi-4-mini-reasoning for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Phi-4-mini-reasoning for details.

Qwen2505bstatic

FoundryModel

This model is an optimized version of Qwen2.5-0.5B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen2.5-0.5B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen2.5-0.5B-Instruct for details.

Qwen2514bstatic

FoundryModel

This model is an optimized version of Qwen2.5-14B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen2.5-14B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen2.5-14B-Instruct for details.

Qwen2515bstatic

FoundryModel

This model is an optimized version of Qwen2.5-1.5B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen2.5-1.5B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen2.5-1.5B-Instruct for details.

Qwen257bstatic

FoundryModel

This model is an optimized version of Qwen2.5-7B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen2.5-7B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen2.5-7B-Instruct for details.

Qwen25Coder05bstatic

FoundryModel

This model is an optimized version of Qwen2.5-Coder-0.5B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen2.5-Coder-0.5B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen2.5-Coder-0.5B-Instruct for details.

Qwen25Coder14bstatic

FoundryModel

This model is an optimized version of Qwen2.5-Coder-14B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen2.5-Coder-14B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen2.5-Coder-14B-Instruct for details.

Qwen25Coder15bstatic

FoundryModel

This model is an optimized version of Qwen2.5-Coder-1.5B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen2.5-Coder-1.5B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen2.5-Coder-1.5B-Instruct for details.

Qwen25Coder3bstatic

FoundryModel

This model is an optimized version of Qwen2.5-Coder-3B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen2.5-Coder-3B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen2.5-Coder-3B-Instruct for details.

Qwen25Coder7bstatic

FoundryModel

This model is an optimized version of Qwen2.5-Coder-7B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen2.5-Coder-7B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen2.5-Coder-7B-Instruct for details.

Qwen306bstatic

FoundryModel

This model is an optimized version of Qwen3-0.6B to enable local inference. This model uses KLD Gradient quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3-0.6B for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3-0.6B for details.

Qwen314bstatic

FoundryModel

This model is an optimized version of Qwen3-14B to enable local inference. This model uses GPTQ quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3-14B for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3-14B for details.

Qwen317bstatic

FoundryModel

This model is an optimized version of Qwen3-1.7B to enable local inference. This model uses KLD Gradient quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3-1.7B for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3-1.7B for details.

Qwen34bstatic

FoundryModel

This model is an optimized version of Qwen3-4B to enable local inference. This model uses KLD Gradient quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3-4B for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3-4B for details.

Qwen3508bstatic

FoundryModel

This model is an optimized version of Qwen3.5-0.8B to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3.5-0.8B for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3.5-0.8B for details.

Qwen352bstatic

FoundryModel

This model is an optimized version of Qwen3.5-2B to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3.5-2B for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3.5-2B for details.

Qwen352bTextstatic

FoundryModel

This model is an optimized text-only version of Qwen3.5-2B to enable local inference with CUDA. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3.5-2B text model for local inference with CUDA.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3.5-2B for details.

Qwen354bstatic

FoundryModel

This model is an optimized version of Qwen3.5-4B to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3.5-4B for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3.5-4B for details.

Qwen359bstatic

FoundryModel

This model is an optimized version of Qwen3.5-9B to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3.5-9B for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3.5-9B for details.

Qwen38bstatic

FoundryModel

This model is an optimized version of Qwen3-8B to enable local inference. This model uses KLD Gradient quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3-8B for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3-8B for details.

Qwen3Embedding06bstatic

FoundryModel

Qwen3 Embedding 0.6B Cuda Gpu

This is the GPU (NVIDIA CUDA)-optimized variant of qwen3-embedding-0.6b, a text embedding model from the Qwen3 family developed by Alibaba Cloud and optimized by Microsoft.

Model Details

• Model Type: Text Embedding (ONNX)

• Parameters: 0.6 billion

• Context Length: 32K tokens

• Embedding Dimension: Up to 1024

• Quantization: KLD Gradient quantization

• Target Device: GPU (NVIDIA CUDA)

• Execution Provider: CUDAExecutionProvider

• Supported Languages: 100+

Intended Use

This model is optimized for local execution on devices with GPU (NVIDIA CUDA) hardware acceleration using Foundry Local.

Capabilities

• Text retrieval and semantic search

• Code retrieval

• Text classification and clustering

• Bitext mining

• Multilingual and cross-lingual retrieval

License

This model is licensed under Apache 2.0. See license details.

Source

• HuggingFace: Qwen3-Embedding-0.6B

Qwen3Embedding8bstatic

FoundryModel

Qwen3 Embedding 8B Cuda Gpu

This is the GPU (NVIDIA CUDA)-optimized variant of qwen3-embedding-8b, a text embedding model from the Qwen3 family developed by Alibaba Cloud and optimized by Microsoft.

Model Details

• Model Type: Text Embedding (ONNX)

• Parameters: 8 billion

• Context Length: 32K tokens

• Embedding Dimension: Up to 4096

• Quantization: KLD Gradient quantization

• Target Device: GPU (NVIDIA CUDA)

• Execution Provider: CUDAExecutionProvider

• Supported Languages: 100+

Intended Use

This model is optimized for local execution on devices with GPU (NVIDIA CUDA) hardware acceleration using Foundry Local.

Capabilities

• Text retrieval and semantic search

• Code retrieval

• Text classification and clustering

• Bitext mining

• Multilingual and cross-lingual retrieval

License

This model is licensed under Apache 2.0. See license details.

Source

• HuggingFace: Qwen3-Embedding-8B

Qwen3Vl2bInstructstatic

FoundryModel

This model is an optimized version of Qwen3-VL-2B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3-VL-2B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3-VL-2B-Instruct for details.

Qwen3Vl4bInstructstatic

FoundryModel

This model is an optimized version of Qwen3-VL-4B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3-VL-4B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3-VL-4B-Instruct for details.

Qwen3Vl8bInstructstatic

FoundryModel

This model is an optimized version of Qwen3-VL-8B-Instruct to enable local inference. This model uses RTN quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the Qwen3-VL-8B-Instruct for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3-VL-8B-Instruct for details.

Smollm33bstatic

FoundryModel

This model is an optimized version of SmolLM3-3B to enable local inference. This model uses GPTQ quantization.

Model Description

• Developed by: Microsoft

• Model type: ONNX

• License: apache-2.0

• Model Description: This is a conversion of the SmolLM3-3B for local inference.

• Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model SmolLM3-3B for details.

View all fields