Overview
SenseNova U1 (RiRiXin U1) is a series of "native understanding and generation unified models" officially released and open-sourced by SenseTime on April 28, 2026. Based on SenseTime's self-developed NEO-unify architecture from March this year, it is the first to natively unify multimodal understanding, reasoning, and generation capabilities within a single model architecture, moving away from the common "stitching" multimodal approach in the industry (i.e., simply stitching together understanding and generation models).
The initial open-source release includes two specifications of the Lite lightweight version: SenseNova-U1-8B-MoT (dense backbone) and SenseNova-U1-A3B-MoT (MoE mixture-of-experts backbone), along with weights, inference code, and technical reports. For developers, this is one of the few domestically open-sourced models in the first half of 2026 with original architectural breakthroughs.
📌 2026 Supplementary Updates:
- 2026/6/9 SenseNova Token Plan Public Beta Launch: All model calls are completely free during the period (DeepSeek V4 Flash + SenseNova 6.7 Flash-Lite + SenseNova U1 Fast) — one of the few "top-tier model free" activities in China
- 2026/7/13 SenseNova-Vision Officially Released and Open-Sourced: An important visual capability upgrade for the RiRiXin system, making vision a native capability of general foundation models, a key complement following U1
- 2026/9/18 SenseNova U1.5 Technical Report Published: An 8B-MoT natively unified multimodal model that understands, reasons about and generates visual content without an external vision encoder or VAE; it improves the visual interface through reconstruction that preserves spatial coherence between patches, trains at up to native 4K, and post-trains separate experts for visual aesthetics, bilingual text rendering, infographic generation and image editing before merging them by multi-expert on-policy distillation. SenseTime open-sourced the training code for supervised fine-tuning, reinforcement learning and on-policy distillation
- 2026/9/21 SenseNova U1 Pro Goes Live: Available in the Xiaohuanxiong app and on the SenseNova API, with up to 8K output and support for unusual aspect ratios. It uses an interleaved text-and-image chain of thought so copy and visual design inform each other across understanding the brief, searching to fill gaps, design generation, editing, and self-checking and correction, aimed at text-heavy work such as infographics, posters, architectural renders and city-themed illustration
Key Features
- NEO-unify Native Unified Architecture: A single model unifies multimodal understanding, reasoning, and generation capabilities, rather than achieving them through multiple model stitching
- Fully Open Source: Weights, code, and technical reports are all open-sourced; enterprises can deploy privately with zero authorization barriers for secondary development
- Dual Backbone Options: The 8B dense version is suitable for low-VRAM deployment, while the A3B MoE version offers stronger capabilities at the same inference cost
- Unified Input/Output for Text, Images, Audio, and Video: Supports native multimodal interaction with images, text, audio, and video without the need for additional adapters
- Optimized for Chinese Scenarios: Specifically tuned for local scenarios such as Chinese understanding, OCR, document parsing, and image generation
- Dual-Channel Release on HuggingFace / GitHub: Weights on HuggingFace, code on GitHub, with ready-to-use inference examples and Docker images
- U1.5 Natively Unified 8B-MoT: Per the 2026/9/18 technical report, an 8B-MoT natively unified multimodal model that understands, reasons about and generates visual content without an external vision encoder or VAE, trains at up to native 4K, and post-trains separate experts for visual aesthetics, bilingual text rendering, infographic generation and image editing before merging them by multi-expert on-policy distillation; training code for supervised fine-tuning, reinforcement learning and on-policy distillation is open-sourced
- SenseNova U1 Pro with Up to 8K Output: Live since 2026/9/21 in the Xiaohuanxiong app and on the SenseNova API, supporting up to 8K output and unusual aspect ratios, with an interleaved text-and-image chain of thought spanning understanding the brief, searching to fill gaps, design generation, editing, and self-checking and correction
Use Cases
- Enterprises and teams needing local/private deployment of multimodal large models
- Developers building multimodal retrieval, visual question answering, and text-to-image generation applications
- Universities and research institutions conducting multimodal architecture research
- Products requiring unified understanding and generation capabilities (e.g., both image understanding and generation)
- Teams seeking to move away from OpenAI/Anthropic commercial APIs for overseas or compliance-sensitive scenarios
- Text-heavy visual work such as infographics, posters, architectural renders and city-themed illustration
Pros
- Original Architecture: NEO-unify is a rare underlying innovation in the domestic open-source ecosystem
- Fully Open Source: Complete set of weights, code, and reports, business-friendly
- Dual Specification Options: 8B suitable for single-GPU deployment, A3B MoE suitable for high-throughput services
- Native Unification: Image viewing, thinking, generation, and description within the same model, no stitching loss
- Friendly to Chinese and OCR scenarios, suitable for domestic business deployment
- Expanding Lineup: U1.5 details a natively unified 8B-MoT trained at up to native 4K, and U1 Pro brings up to 8K output with an interleaved text-and-image chain of thought
Pricing
The SenseNova U1 Lite series (SenseNova-U1-8B-MoT, SenseNova-U1-A3B-MoT) is open-sourced and free to use, with no licensing fees for commercial scenarios. SenseTime also provides a hosted API on the RiRiXin official platform, billed per token, with specific pricing subject to official announcements; SenseNova U1 Pro is likewise available on the SenseNova API. For the MoE version, it is recommended to use mainstream inference frameworks (vLLM, SGLang, etc.) for optimal throughput.
Summary
SenseNova U1 is a heavyweight new member of the domestic open-source multimodal model landscape in spring 2026. Its biggest highlight is the NEO-unify native unified architecture—integrating understanding and generation into the same model rather than stitching them together. For teams needing multimodal capabilities but wanting to reduce reliance on overseas APIs, it offers a new option that is both open-source and architecturally original, especially suitable for Chinese scenarios and private deployment. The lineup has since expanded: the 2026/9/18 U1.5 technical report details an 8B-MoT natively unified model that works without an external vision encoder or VAE and trains at up to native 4K, while the 2026/9/21 SenseNova U1 Pro brings up to 8K output and an interleaved text-and-image chain of thought to the Xiaohuanxiong app and the SenseNova API. For those seeking top-tier overall performance, Claude/GPT series can still be used as cloud supplements.
Version History
- SenseNova U1 Pro goes live with up to 8K output (2026-09-21): SenseTime released SenseNova U1 Pro, available in the Xiaohuanxiong app and on the SenseNova API. It uses an interleaved text-and-image chain of thought so copy and visual design inform each other across understanding the brief, searching to fill gaps, design generation, editing, and self-checking and correction. It supports up to 8K output and unusual aspect ratios, aimed at text-heavy work such as infographics, posters, architectural renders and city-themed illustration.
- SenseNova U1.5 technical report published (2026-09-18): SenseTime published the SenseNova U1.5 technical report for an 8B-MoT natively unified multimodal model that understands, reasons about and generates visual content without an external vision encoder or VAE. It improves the visual interface through reconstruction that preserves spatial coherence between patches, trains at up to native 4K, and post-trains separate experts for visual aesthetics, bilingual text rendering, infographic generation and image editing before merging them by multi-expert on-policy distillation. SenseTime open-sourced the training code for supervised fine-tuning, reinforcement learning and on-policy distillation.