# Is FineART-VLA SOTA?

Pending benchmarks. Announced Oct 29, 2026; no independent results yet.

As of Nov 3, 2026, FineART-VLA is pending benchmarks on isitsota: announced Oct 29, 2026, with no independent results on the boards we track yet.

An open vision-language-action (VLA) robot model from Scale AI and Hugging Face's LeRobot team that plans before it acts: a System 2 part predicts the next subtask in words and a System 1 part turns it into arm actions. Fine-tuned from Physical Intelligence's pi0.5 on FineART, Scale's dataset of 40,543 two-arm robot episodes with 533,913 subtask labels. About 4.1B parameters; weights on Hugging Face under the Gemma license.

Canonical: https://isitsota.com/m/fineart-vla
Last updated: 2026-10-29
Tracking since: 2026-10-29

## Announced facts

| Fact | Value |
| --- | --- |
| Lab | Scale AI (Scale Labs) with Hugging Face's LeRobot team; published in the lerobot Hugging Face org |
| Announced | Oct 29, 2026 |
| API | Not yet |
| Open weights | Published · Hugging Face, about 9.4 GB (F32 and BF16 safetensors) under the Gemma license inherited from PaliGemma/pi0.5; LeRobot code is Apache-2.0 |
| Parameters | About 4.1B (4,143,404,816 in the safetensors): a Gemma 2B vision-language backbone plus a Gemma 300M action expert |
| Type | Vision-language-action robot policy, not a chat model: camera frames, a language instruction and robot state in, actions out |
| Design | System 2 predicts subtasks in natural language; System 1 turns the current subtask into actions (50-step action chunks, 10 flow-matching steps) |
| Base model | lerobot/pi05_base, LeRobot's port of Physical Intelligence's pi0.5 (PaliGemma-based) |
| Robot | Stationary two-arm ALOHA-style setup: three cameras (high, left wrist, right wrist) at 224×224, 14-dimensional state and action |
| Training data | FineART dataset (CC-BY-4.0): 40,543 episodes, 1,718 hours, 151 tasks, 533,913 subtask annotations (about 13 per episode) |
| Code | Training and fine-tuning through LeRobot (lerobot-train; policy fineart_vla); lerobot-annotate labels subtasks on other LeRobot datasets |
| Paper | arXiv 2609.36416, first posted Sep 29, 2026 (UTC); the weights followed on Oct 29, 2026 |

Tags marked unsure are not stated on the lab's own pages.

## What would let us call it

Verdicts use only independent boards we already track. Two consecutive snapshots are required before any crown.

- [x] Publicly announced with a source
- [x] Usable by the public (API or weights)
- [ ] Listed on an independent board we track
- [ ] Enough results for a rank in a category
- [ ] Holds across two snapshots (crowns only)

## What the lab claims

Self-reported. Shown for context; not used in any verdict.

- Lab claim: FineART is the largest open robot dataset with subtask labels.
- Lab claim: across 3,400 real-world evaluations, subtask training lifts spatial-instruction success from 32.0% to 100.0% and, with step-by-step guidance, unseen long-horizon task success from 16.0% to 76.0%.
- Lab claim: on a new robot (YAM arms) it matches the baseline with 10x less fine-tuning data.

## Sources

- LAB: [Scale Labs blog: FineART, Scale AI's first open-source robotics dataset and model](https://labs.scale.com/blog/fineart) (2026-10-29)
- LAB: [Scale Labs announcement on X](https://x.com/ScaleAILabs/status/2115837725719048535) (2026-10-29)
- REPO: [lerobot/FineART-VLA-subtask-midtrained on Hugging Face](https://huggingface.co/lerobot/FineART-VLA-subtask-midtrained)
- REPO: [lerobot/FineART dataset on Hugging Face](https://huggingface.co/datasets/lerobot/FineART)
- LAB: [LeRobot docs: FineART-VLA](https://huggingface.co/docs/lerobot/main/fineart_vla)
- PAPER: [FineART paper (arXiv 2609.36416)](https://arxiv.org/abs/2609.36416)

Tracking continues until an independent board lists it.

## Related

- Lab: [Scale AI](https://isitsota.com/labs/scale-ai.md)
- All pending releases: https://isitsota.com/releases.md
- JSON: https://api.isitsota.com/v1/models/fineart-vla