# RynnBrain-2B by DAMO Academy: Open-Weight Model Source: https://savrn.com/models/rynnbrain-2b Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve RynnBrain-2B (2.4B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them. | Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits | | --- | --- | --- | --- | --- | --- | | 16-bit | 4.9 GB | 5.9 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 2.4 GB | 2.9 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 1.2 GB | 1.5 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the [SAVRN Index](https://savrn.com/ai-index/pricing/gpus), read Oct 23, 2026. [RynnBrain-2B on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/rynnbrain-2b/gpus) ## Model Card By DAMO Academy, published under apache-2.0, revision ad71236d013a. RynnBrain: Open Embodied Foundation Models If you like our project, please give us a star on Github for the latest update. --- ## News - **[2026.02.02]** Release **RynnBrain family** weights and inference code. - **[2026.02.02]** Add cookbooks for cognition, localization, reasoning, and planning. ## Introduction RynnBrain aims to serve as a **physics-aware embodied brain**: it observes egocentric scenes, grounds language to physical space and time, and supports downstream robotic systems with reliable localization and planning outputs. ### Key Highlights - **Comprehensive egocentric understanding** Strong spatial comprehension and egocentric cognition across embodied QA, counting, OCR, and fine-grained video understanding. - **Diverse spatiotemporal localization** Locates objects, target areas, and predicts trajectories across long episodic context, enabling global spatial awareness. - **Physical-space grounded reasoning (RynnBrain family)** The broader RynnBrain family includes “Thinking” variants that interleave textual reasoning with spatial grounding to anchor reasoning in reality. - **Physics-aware precise planning (RynnBrain family)** Integrates localized affordances/areas/objects into planning outputs to provide downstream VLA models with… [Read the full model card (527 words)](https://savrn.com/models/rynnbrain-2b/card) ## Configuration Architecture Qwen3VLForConditionalGeneration Context length (tokens) 262,144 Layers 28 Hidden size 2,048 Feed-forward size 6,144 Attention heads 16 Key/value heads 8 Head dimension 128 Vocabulary size 151,936 RoPE base 5,000,000 Model type qwen3_vl ## Identity and Version Repository Alibaba-DAMO-Academy/RynnBrain-2B Publisher DAMO Academy Task Image and text to text Modality Image and text Library transformers Parameters 2.4B parameters Languages en, zh Revision ad71236d013ae5618833fded73a32569045a8e9e First published 2026-01-31 Last updated 2026-04-15 ## Files and Weights 14 files, 4.9 GB in total. The weights are 1 file totalling 4.9 GB in safetensors. Weights1 file · 4.9 GB Configuration6 files · 4.7 KB Tokenizer4 files · 15.9 MB Documentation1 file · 7.4 KB Other1 file · 5.3 KB Repository1 file · 1.6 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 4.9 GB | ac4ae115b483 | | added_tokens.json | Configuration | 707 B | — | | config.json | Configuration | 1.6 KB | — | | generation_config.json | Configuration | 199 B | — | | preprocessor_config.json | Configuration | 782 B | — | | special_tokens_map.json | Configuration | 613 B | — | | video_preprocessor_config.json | Configuration | 817 B | — | | README.md | Documentation | 7.4 KB | — | | chat_template.jinja | Other | 5.3 KB | — | | .gitattributes | Repository | 1.6 KB | — | | merges.txt | Tokenizer | 1.7 MB | — | | tokenizer.json | Tokenizer | 11.4 MB | aeb13307a71a | | tokenizer_config.json | Tokenizer | 5.4 KB | — | | vocab.json | Tokenizer | 2.8 MB | — | ## License and Download License apache-2.0 Access Open weights, no gate Download size 4.9 GB [Download from DAMO Academy](https://huggingface.co/Alibaba-DAMO-Academy/RynnBrain-2B) Released by DAMO Academy through its official repository on Hugging Face. [Read the license](https://www.apache.org/licenses/LICENSE-2.0). ## Built From - Described by arXiv:2602.14979 ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 4.9 GB | | 16-bit | 4.9 GB | | 8-bit | 2.4 GB | | 4-bit | 1.2 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About RynnBrain-2B ### How much GPU memory does RynnBrain-2B need? About 5.9 GB at 16-bit and 1.5 GB at 4-bit: the weights (2.4B parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run RynnBrain-2B on? At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists. ### Can I use RynnBrain-2B commercially? Yes. RynnBrain-2B is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors. ### What is RynnBrain-2B's context length? 262,144 tokens, from the maximum position embeddings in its published configuration. ## Similar Models Model · Image and text to text ### [Cosmos-Reason2-2B](https://savrn.com/models/cosmos-reason2-2b) [NVIDIA](https://savrn.com/model-publishers/nvidia) NVIDIA Cosmos Reason 2 is an open, customizable, 2B-parameter reasoning vision language model (VLM) for physical AI and robotics that enables robots and vision AI agents to reason like humans, using prior knowledge, physics understanding and common sense to understand and act in the real world. This model understands space, time, and fundamental physics, and can serve as a planning model to reason what steps an embodied agent might take next. New features with Cosmos Reason 2: Enhanced physical AI reasoning with improved spatio-temporal understanding and timestamp precision. Supports object detection with 2D/3D point localization and bounding box coordinates with reasoning explanations and… Access requested at publisher other 2.4B parameters cosmos [View model](https://savrn.com/models/cosmos-reason2-2b) Model · Image and text to text ### [ProcVLM-2B](https://savrn.com/models/procvlm-2b) [Youhe Feng](https://savrn.com/model-publishers/ce-amtic) ProcVLM-2B is a procedure-grounded vision-language model for estimating progress rewards from robot manipulation observations. Given a task description and a recent window of video frames, the model reasons about the remaining atomic actions and predicts the current task completion percentage. ProcVLM-2B is designed for research on robot learning, progress reward modeling, embodied evaluation, and procedure-aware video understanding. Typical use cases include: - estimating task completion progress from robot videos; - producing dense progress rewards from sparse demonstrations; - adapting progress prediction to a new environment with one-shot LoRA fine-tuning. This model is not intended to… Open weights cc-by-4.0 2.4B parameters 262,144 tokens transformers [View model](https://savrn.com/models/procvlm-2b) Model · Image and text to text ### [InternVL3.5-2B-MMBU-RFT](https://savrn.com/models/internvl3-5-2b-mmbu-rft) [Nitiz Khanal](https://savrn.com/model-publishers/staratnyte1) A full fine-tune of OpenGVLab/InternVL35-2B-HF (vision tower, projector and LLM all trained) on rejection-sampled, verified answers from public biomedical datasets. No MMBU data was used, and every training image passes a pHash near-duplicate check against the MMBU public images. - RFT data. The base model answered public classification, fine-grained, detection and counting questions 6 times each. Only answers that verified against the ground truth were kept. Hint-rationalized answers were added for unsolved questions, and box-answer rows were excluded. - Metadata supervision (about 30% of the mix). Targets are modality, submodality, specimen, body part, stain and medical domain. Medical… Open weights apache-2.0 2.3B parameters 40,960 tokens [View model](https://savrn.com/models/internvl3-5-2b-mmbu-rft) Model · Image and text to text ### [Qwen3.8-27B-CQ-4bit](https://savrn.com/models/qwen3-8-27b-cq-4bit) [ThousandML](https://savrn.com/model-publishers/thousandml) Qwen3.8 27B quantized to 4-bit with Coded Quantization (CQ). CQ is a calibration-free 4-bit quantization method designed to preserve the behavior of the original model without requiring a calibration dataset. Because CQ does not optimize the quantization around a fixed calibration corpus, it is designed to generalize beyond calibration-specific data and workloads. This repository contains the 27B-parameter Qwen3.8 model quantized with Coded Quantization (CQ) 4-bit. Lower mean KL divergence indicates closer agreement between the quantized model's output distribution and the original model. Top-1 agreement measures how often the quantized model and original model select the same… Open weights apache-2.0 2.6B parameters 262,144 tokens [View model](https://savrn.com/models/qwen3-8-27b-cq-4bit) Model · Image and text to text ### [Qwen3.5-2B](https://savrn.com/models/qwen3-5-2b) [Qwen](https://savrn.com/model-publishers/qwen) Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Qwen3.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to empower developers and enterprises with unprecedented capability and efficiency. For more details, please refer to our blog post Qwen3.5. WMT24++: a harder subset of WMT24 after difficulty labeling and rebalancing; we report the averaged scores on 55 languages using XCOMET-XXL. Empty cells (--) indicate scores not yet available or not applicable. Scores of Qwen3.5 models are reported… Open weights apache-2.0 2.3B parameters 262,144 tokens transformers [View model](https://savrn.com/models/qwen3-5-2b) Model · Image and text to text ### [Rax-4.5](https://savrn.com/models/rax-4-5) [RaxCore](https://savrn.com/model-publishers/raxcore-dev) Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Rax 4.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to empower developers and enterprises with unprecedented capability and efficiency. Rax 4.5 features the following enhancement: For more details, please refer to our blog post Rax 4.5. WMT24++: a harder subset of WMT24 after difficulty labeling and rebalancing; we report the averaged scores on 55 languages using XCOMET-XXL. Empty cells (--) indicate scores not yet available or not… Open weights apache-2.0 2.3B parameters 262,144 tokens transformers [View model](https://savrn.com/models/rax-4-5) ## DAMO Academy [All models and datasets](https://savrn.com/model-publishers/alibaba-damo-academy) ## Versions - [ad71236d013a](https://savrn.com/models/rynnbrain-2b/versions/ad71236d013a) · current 2026-09-18 ## Explore More - [All image and text to text models](https://savrn.com/models/tasks/image-and-text-to-text) - [All models under apache-2.0](https://savrn.com/models/licenses/apache-2-0) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-21. - [Hugging Face record](https://huggingface.co/Alibaba-DAMO-Academy/RynnBrain-2B) - [How the hub is built](https://savrn.com/model-hub/methodology)