LIVE

Publisher profile

lmstudio-community AI models

Explore active AI models attributed to lmstudio-community across indexed source catalogs. Open any profile for capabilities, context, pricing and verified source links.

Models
10
Categories
1
Sources
1
vision

Available catalog

Published models

10 active

Hugging Face↗

gemma-4-31B-it-MLX-4bit

gemma-4-31B-it-MLX-4bit is an AI model record indexed from Hugging Face.

visionimagetext
Size
31B parameters
Context
Not listed
Hugging Face↗

gemma-4-31B-it-MLX-5bit

gemma-4-31B-it-MLX-5bit is an AI model record indexed from Hugging Face.

visionimagetext
Size
31B parameters
Context
Not listed
Hugging Face↗

gemma-4-31B-it-MLX-6bit

gemma-4-31B-it-MLX-6bit is an AI model record indexed from Hugging Face.

visionimagetext
Size
31B parameters
Context
Not listed
Hugging Face↗

gemma-4-31B-it-MLX-8bit

gemma-4-31B-it-MLX-8bit is an AI model record indexed from Hugging Face.

visionimagetext
Size
31B parameters
Context
Not listed
Hugging Face↗

Laguna-S-2.1-GGUF

Laguna-S-2.1-GGUF is an AI model record indexed from Hugging Face.

Size
Size not listed
Context
Not listed
Hugging Face↗

Laguna-XS-2.1-GGUF

Laguna-XS-2.1-GGUF is an AI model record indexed from Hugging Face.

Size
Size not listed
Context
Not listed
Hugging Face↗

Qwen3.8-27B-MLX-4bit

Qwen3.8-27B-MLX-4bit is a Hugging Face repository for an image-text-to-text model that accepts text and image inputs. It is therefore suited to vision-oriented applications such as analyzing images alongside written prompts and producing text responses. The repository description identifies the model as having 27B parameters, while the recorded exact parameter count is 27,356,728,560. Its name indicates an MLX 4-bit variant, which may be relevant when selecting a compatible runtime, but the supplied material does not provide runtime instructions, hardware requirements, memory usage, or an inference-speed benchmark. No advertised context-window value, pricing information, or additional capability labels are available in the supplied material. The source identifies the repository as licensed under Apache-2.0, subject to the terms of that license. Developers should verify compatibility with their intended MLX deployment environment and validate image understanding quality for their specific workloads, since benchmark results and detailed model-card guidance are not included here. This entry is best considered a multimodal vision model listing rather than a complete performance or deployment specification.

visiontextimage
Size
27.4B parameters
Context
Not listed
Hugging Face↗

Qwen3.8-27B-MLX-5bit

Qwen3.8-27B-MLX-5bit is a 27.4-billion-parameter image-text-to-text model repository on Hugging Face, intended for applications that combine visual and textual input. Its listed modalities are text and image, and the supplied material identifies vision as its use case. A developer could evaluate it for image understanding, visual question answering, and workflows that require generating text from image and text inputs, although the excerpt does not provide task-specific examples or benchmark results. The repository is described as being available under the Apache-2.0 license, which is generally permissive, but deployment should still be checked against the repository’s current license text and any associated usage conditions. An advertised context window is not available in the supplied material, and no runtime defaults are provided. Inference speed, pricing, and performance comparisons are also unavailable. The model name identifies an MLX 5-bit variant, but the supplied material does not document hardware requirements, supported runtimes, quantization trade-offs, or production-readiness details. Review the repository files before integrating it into an application.

visiontextimage
Size
27.4B parameters
Context
Not listed
Hugging Face↗

Qwen3.8-27B-MLX-6bit

Qwen3.8-27B-MLX-6bit is a Hugging Face repository for an image-text-to-text model intended for applications that combine visual and textual input. The supplied model metadata lists text and image modalities, and the repository description identifies it as an image-text-to-text model with 27B parameters. Its listed practical use case is vision, making it potentially relevant for image understanding and generating text from image-and-text prompts. The model record reports 27,356,728,560 parameters for this exact model. The repository is identified under the Apache-2.0 license, which is the primary licensing detail available for evaluating integration and redistribution; developers should still review the repository’s complete license terms before deployment. The supplied material does not provide an advertised context window, inference speed benchmark, pricing, or a detailed capability list. It also does not specify supported serving software, hardware requirements, input limits, or quality benchmarks. As a result, this entry supports identifying the model’s modality and scale, but additional testing and repository review are needed to assess runtime compatibility, latency, memory requirements, and performance for a specific vision workflow.

visiontextimage
Size
27.4B parameters
Context
Not listed
Hugging Face↗

Qwen3.8-27B-MLX-8bit

Qwen3.8-27B-MLX-8bit is a reasonable candidate for developers seeking a large multimodal model repository focused on image-and-text-to-text tasks. The supplied Hugging Face metadata describes it as an Image-Text-to-Text model with 27B parameters, and the record identifies text and image as its modalities. Its primary listed use case is vision, so it may be relevant for applications that interpret images and produce textual responses, provided the intended runtime supports the repository’s MLX-8bit format indicated by the model name. The exact parameter count is 27,356,728,560. The Apache-2.0 license makes it suitable for many use cases, subject to the license’s terms and any obligations applicable to your deployment. The supplied material does not provide an advertised context window, inference throughput benchmark, pricing, or a detailed capabilities list. It also does not include enough information to assess image resolution limits, quantization quality, hardware requirements, tool-calling support, coding performance, safety behavior, or production reliability. Developers should therefore validate compatibility, memory requirements, output quality, and licensing fit in their own environment before adoption.

visiontextimage
Size
27.4B parameters
Context
Not listed

Continue exploring

Related guides and tools