Alibaba DAMO AcademyChina2026

RynnBrain

An AI model: a vision-language-action model (VLA) takes camera images and an instruction and turns them into movements for a robot. This page describes the software, not a robot.

Robotsnot stated

VersionsRynnBrain 1.0 2026-02-09RynnBrain 1.1 2026-07-16

The developer has released this model in several versions. Each figure applies to the version its source names.

Key figures as stated by the developer

4 of 6 stated · 2 gaps

Model type ⁠B embodied reasoning
Controls not stated
Inputs ⁠A camera language
Access ⁠A open download
Parameters not stated
Base model ⁠D Qwen3.5

How to read the figures

14kgA30.9lbA: The letter in the grey square after a figure points to the source the figure came from. Robots with several sources carry several letters.
74.5lb: Imperial figures are our own conversion unless the manufacturer states an imperial figure itself. The source letter always belongs to the metric figure — a conversion has no source of its own.

The four states

Versions and notes in the key figures (6)

Model type

Developer wording: "specializing in diverse, fine-grained embodied reasoning and complex planning tasks".

Controls

Note RynnBrain 1.1 README: "It encodes omni-vision inputs with language instructions and produces aligned outputs — text, pointing sequences, 3D perception, and contact signals".

Inputs

RynnBrain 1.1 README: "It encodes omni-vision inputs with language instructions and produces aligned outputs — text, pointing sequences, 3D perception, and contact signals".

Access

RynnBrain 1.1 2026-07-16: "Release RynnBrain 1.1 checkpoints (2B / 9B / 122B-A10B)" RynnBrain 1.0 2026-02-09: "Release our code and model checkpoints!!"

Parameters

Note The developer gives one number per size, never one for RynnBrain. RynnBrain 1.1 2026-07-16: "RynnBrain 1.1 is released in three scales: 2B, 9B, and 122B-A10B". RynnBrain 1.0 2026-02-09: "RynnBrain is available in three dense variants (2B, 4B, and 8B) and one mixture-of-experts (MoE) model (30B-A3B)."

Base model

RynnBrain 1.1 2026-07-16: "a large language model (LLM) backbone initialized from Qwen3.5 variants". RynnBrain 1.0 2026-02-09: Qwen3-VL-Instruct models, for example "base_model: Qwen3-VL-8B-Instruct".

Versions

RynnBrain 1.0 2026-02-09: Qwen3-VL-Instruct
RynnBrain 1.1 2026-07-16: Qwen3.5

Full specifications

5 of 12 fields stated

The letter points to the source at the foot of the page. Fields without a figure are dashed: the developer does not state them — that is not the same as zero, and not the same as no.
Field	Value	Source and caveats
Input and output
Model type	`embodied reasoning`	B Developer wording: "specializing in diverse, fine-grained embodied reasoning and complex planning tasks".
Inputs	`camera` `language`	A RynnBrain 1.1 README: "It encodes omni-vision inputs with language instructions and produces aligned outputs — text, pointing sequences, 3D perception, and contact signals".
Controls	not stated	Note RynnBrain 1.1 README: "It encodes omni-vision inputs with language instructions and produces aligned outputs — text, pointing sequences, 3D perception, and contact signals". ⁠A
Degrees of freedom controlled	not stated
Size and speed
Parameters	not stated	Note The developer gives one number per size, never one for RynnBrain. RynnBrain 1.1 2026-07-16: "RynnBrain 1.1 is released in three scales: 2B, 9B, and 122B-A10B". RynnBrain 1.0 2026-02-09: "RynnBrain is available in three dense variants (2B, 4B, and 8B) and one mixture-of-experts (MoE) model (30B-A3B)." ⁠A
Control rate	not stated
Where it runs and what it is built on
Runs where	not stated
Base model	Qwen3.5	D RynnBrain 1.1 2026-07-16: "a large language model (LLM) backbone initialized from Qwen3.5 variants". RynnBrain 1.0 2026-02-09: Qwen3-VL-Instruct models, for example "base_model: Qwen3-VL-8B-Instruct". Versions RynnBrain 1.0 2026-02-09 Qwen3-VL-Instruct RynnBrain 1.1 2026-07-16 Qwen3.5
Openness
Access	open download	A RynnBrain 1.1 2026-07-16: "Release RynnBrain 1.1 checkpoints (2B / 9B / 122B-A10B)" RynnBrain 1.0 2026-02-09: "Release our code and model checkpoints!!"
License	Apache 2.0	C RynnBrain 1.1 model card header: "license: apache-2.0". The repository README says "This project is licensed under the Apache License 2.0".
The developer’s own claims
Learns new tasks	not stated
Works across robots	not stated	Yes only if the developer itself says the model runs on different robot models, for example robots from two manufacturers or two of one manufacturer’s models. Several units of the same robot do not count.

Notes

Alibaba DAMO Academy describes RynnBrain as an open-source spatiotemporal foundation model for embodied intelligence.
RynnBrain 1.1 is described by its developer as a systematic upgrade of RynnBrain; RynnBrain 1.0 is still published on a separate branch of the repository.
RynnBrain 1.0 was released on 9 February 2026 and RynnBrain 1.1 on 16 July 2026.

Sources

A Manufacturer product page. github.com/alibaba-damo-academy/RynnBrain · Retrieved 2026-10-13
B github.com/alibaba-damo-academy/RynnBrain/tree/rynnbrain1.0 · Retrieved 2026-10-13
C huggingface.co/Alibaba-DAMO-Academy/RynnBrain1.1-2B · Retrieved 2026-10-13
D arxiv.org/html/2607.17977 · Retrieved 2026-10-13

One line per source, not per figure: a source has one retrieval date. The letter beside a figure points to the line it came from.