# Robbyant/lingbot-vla

A Pragmatic VLA Foundation Model

Repository: https://github.com/Robbyant/lingbot-vla
Canonical: https://ross.abutalabs.com/products/lingbot-vla
Language: Python
License: Apache-2.0
License Family: permissive
Topics: embodied-ai, vla
Last push: 2026-06-11T16:51:17+00:00

## Health v2 (maintenance only)
Score: 54/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 87, release rhythm 35, longevity 15
- inputs: {"age_days": 219, "days_push": 83, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1777, forks 183 (observed 2026-08-28T04:05:34.755942+00:00)

## What it is
LingBot-VLA is a Vision-Language-Action foundation model for robot manipulation, pretrained on 20,000 hours of real-world dual-arm robot data. It includes training and inference code, pretrained checkpoints (with and without depth), and efficient post-training support for custom robot setups.

## Use cases
- train a VLA model for dual-arm robot manipulation
- post-train a pretrained VLA foundation model on my own robot data
- run robot policy inference from camera images and language instructions
- fine-tune a vision-language-action model efficiently on GPU
- evaluate a robot policy in simulation benchmarks
- get pretrained VLA weights for a bimanual robot arm

## When to choose
- you need a strong VLA base model for real-robot manipulation
- you want faster VLA training than existing codebases
- you work with dual-arm robot configurations supported by its pretraining data

## When to avoid
- you need a general-purpose LLM or chatbot
- you have no robot hardware or simulation setup
- you need non-PyTorch or CPU-only training

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, llm-training, robotics
- domain: artificial-intelligence, machine-learning, robotics, deep-learning
- platform: python
- tags: vla, vision-language-action, foundation-model, embodied-ai, robot-learning, post-training, lerobot, linux, gpu

## Member repositories
- Robbyant/lingbot-vla (main) score 54

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:34.755942+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:25:12.115105+00:00, confidence not recorded.
  - readme: https://github.com/Robbyant/lingbot-vla (fetched 2026-08-28T04:05:34.755942+00:00, sha 29d2ed2e9f0a)
- Data as of 2026-08-30T08:39:29.467469+00:00.