trl-internal-testing/tiny-LlavaForConditionalGeneration is a minimal test model for Llava-style vision-language conditional generation. It is used by the TRL (Transformers Reinforcement Learning) team to test training, inference, and chat template functionality for multimodal models. The model supports image and text inputs.
Tiny LlavaForConditionalGeneration sits in PulseGate's Multimodal & vision category. It focuses on enabling unit tests for multimodal (vision-language) training and inference pipelines within the TRL library. Tiny LlavaForConditionalGeneration is an open-source project aimed at developers. Tiny LlavaForConditionalGeneration is open source under the Apache-2.0 license. It ships for the web, the command line, and API.
It is developed by TRL Internal Testing, and it first shipped in 2020. The project is developed in the open on GitHub with 18.9k stars and 521 commits in the last 90 days. Among its 3 catalogued features are Multimodal Generation, Test Model, and vision-Language.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do