The Best OpenVLA Alternatives in 2026

DeepSeek Coder is a language model project for code generation and programming tasks.

DeepSWE benchmarks coding agents on original, long-horizon software engineering tasks.

MobiLlama is a small language model designed for edge devices, according to its repository description.

This repository contains the DeepSeek-V3 language model project and related source files.

ChatGLM-6B is an open bilingual dialogue language model described in English and Chinese.

xTuring supports personalizing open source language models through preprocessing and fine tuning.

This repository contains Google Research code for the Flan-T5 language model project.

Janus combines multimodal understanding and image generation models in a unified model series.

Point·E is a research project that generates 3D point clouds using diffusion models.

CodeGen provides open-source models for program synthesis and code generation.

Code and documentation to train Stanford's Alpaca models, and generate the data.

timm provides PyTorch image encoders and backbones along with training, evaluation, inference, and export scripts.

The learn2learn library supports meta learning experiments and research with PyTorch.

MetaWorld provides robotics environments for benchmarking multi task and meta reinforcement learning.

This repository accompanies research on improving language model factuality and reasoning through multi-agent debate.

WizardLM collects language models developed with Evol-Instruct, including WizardCoder and WizardMath.

Qwen3-Coder provides a code-focused model in the Qwen family of large language models.

The Qwen3 model series was developed by the Qwen team at Alibaba Cloud.

Alibi explains machine learning model predictions through a set of algorithms.

Use SmolVLA, a vision-language-action model trained on data from the LeRobot community, for robotics research.

ImageBind explores a shared embedding space that connects different kinds of data.

This repository implements a GPT-4 agent for playing imperfect information games with theory of mind.

DeepSeek-VL2 supports multimodal understanding through a mixture-of-experts vision-language model.

Bittensor is an open machine intelligence network where subnets produce digital commodities such as compute and inference, with contributors earning TAO.