ERNIE-Wizshot-ME
An introduction in this language is coming soon.
MODEL DIRECTORY
New releases, in one place. Explore models by their publication date.
An introduction in this language is coming soon.
ERNIE-Light-DT-1.0 is a lightweight, long-context text-to-text model available through Baidu AI Cloud’s Qianfan platform. Baidu’s November 12, 2026 announcement describes ERNIE Light Deep Thinking as…
ERNIE 5.5 Flash Preview is a preview foundation model and the first post-trained model built on ERNIE 5.5 Flash Base. Baidu announced it on November 11, 2026. It has 280 billion total parameters,…
ERNIE 5.5 Flash Base is a text-only, sparse mixture-of-experts base language model released on November 5, 2026. It is the checkpoint at the end of pre-training, before mid-training or post-training.…
Baidu released and open-sourced 文心 Unlimited OCR 2 on October 27, 2026. It is a multimodal document-parsing model designed for long, multi-page documents. Baidu’s Qianfan documentation describes it as…
Baidu Flex is an end-to-end speech-language model for flexible, task-oriented real-time voice interactions. Its public beta was announced on September 16, 2026. (cloud.baidu.com) Understands speech…
HPD-Parsing is a lightweight, approximately 1-billion-parameter vision-language model for high-throughput document parsing. The published model repository is PaddlePaddle/HPD-Parsing; this is a…
Unlimited-OCR is Baidu’s open-weight vision-language model for OCR and long-document parsing. The original model was released on June 22, 2026. It is distinct from Unlimited-OCR 2, released later.…
PP-OCRv6_medium_rec is the medium-tier text recognition component of the PP-OCRv6 OCR family. It converts images of text into recognized text; it is not a general-purpose conversational model. The…
PP-OCRv6_medium_det is the medium-scale text detection model in Baidu's PP-OCRv6 family, intended primarily for server deployment. It uses a PPLCNetV4 backbone and RepLKFPN, and its published model…
PP-OCRv6_small_rec is the small-tier text recognition component of Baidu’s PP-OCRv6 OCR family, intended to balance accuracy and efficiency for mobile and desktop deployment. It uses a PPLCNetV4…
PP-OCRv6_small_det is the small text-detection model in PaddleOCR’s PP-OCRv6 series. It locates text in images; it is not a standalone text-transcription or generative language model. The model has…
PP-OCRv6_tiny_rec is the lightweight text-recognition component of PaddleOCR’s PP-OCRv6 family, intended for resource-constrained edge and IoT deployment. It uses a PPLCNetV4 backbone with LightSVTR…
PP-OCRv6_tiny_det is the ultra-lightweight text-detection model in the PP-OCRv6 family, intended for edge and IoT deployment. It has approximately 0.43 million parameters and a listed model size of 1.…
NAVA (Native Audio-Visual Alignment for Generation) is Baidu ERNIE Team’s 6.3-billion-parameter model for jointly generating synchronized audio and video. Its Align-then-Fuse MMDiT architecture first…
PaddleOCR-VL-1.6 is a compact, 0.9-billion-parameter vision-language model for document parsing, upgraded from PaddleOCR-VL-1.5. It uses region-aware data optimization and progressive post-training,…
ERNIE 5.1 is a general-purpose model derived from ERNIE 5.0, positioned for text generation, reasoning, knowledge-intensive work and agentic tasks. Baidu officially announced it on May 9, 2026. (ernie…