What is the context length of ERNIE 4.5 21B A3B Thinking?

ERNIE 4.5 21B A3B Thinking has a 131,072 token context window.

What other text models does Baidu have?

ERNIE-5.5-Flash-Preview (free) and ERNIE-Light-DT-1.0 are other text models from Baidu.

When was ERNIE 4.5 21B A3B Thinking released?

ERNIE 4.5 21B A3B Thinking was released on October 9, 2025. Its knowledge cutoff is March 31, 2025.

Baidu: ERNIE 4.5 21B A3B Thinking

baidu/ernie-4.5-21b-a3b-thinking

Model weights

ERNIE-4.5-21B-A3B-Thinking is Baidu's upgraded lightweight MoE model, refined to boost reasoning depth and quality for top-tier performance in logical puzzles, math, science, coding, text generation, and expert-level academic benchmarks.

Modalities

Context

131K

Released

Oct 9, 2025

Knowledge cutoff

Mar 2025

Not currently available through EU in-region routing. In-region routing docs

More models from baidu

ERNIE-5.5-Flash-Preview (free)

ERNIE 5.5 Flash Preview is the latest foundation model in the ERNIE family, with comprehensive improvements in coding, agentic capabilities, knowledge work, and deep research. It has 280 billion total parameters, activates 14 billion parameters per token, and supports a context window of 1 million tokens. Its new architecture combines strong model capabilities with efficient inference, delivering generation speeds of more than 200 tokens per second.

Text1.0M contextFree

ERNIE-Light-DT-1.0

ERNIE-Light-DT-1.0 is based on open-source GLM models and includes research findings from Baidu Search's Lightning large model. It features Baidu Sparse Attention (BSA), a unique sparse attention mechanism developed by Baidu ERNIE. By using feedback-driven history compression, it improves the storage and indexing of historical information. This significantly reduces KV cache storage and computational costs while maintaining model performance. As a result, it allows for longer context windows at a lower cost.

Text1.0M context$1.13 / $3.94

CoBuddy

CoBuddy is a code generation model from Baidu, optimized for coding tasks and AI Agent workflows. It features high inference throughput and low end-to-end latency, with native support for tool calling and reasoning. The model runs on fp8 quantization with a 131K token context window and up to 65K output tokens.

Text131K context

Qianfan-OCR-Fast

Qianfan-OCR-Fast is a domain-specific multimodal large model purpose-built for OCR. By leveraging specialized OCR training data while preserving versatile multimodal intelligence, it provides a powerful performance upgrade over Qianfan-OCR.

Text66K context

ERNIE 4.5 21B A3B

A sophisticated text-based Mixture-of-Experts (MoE) model featuring 21B total parameters with 3B activated per token, delivering exceptional multimodal understanding and generation through heterogeneous MoE structures and modality-isolated routing. Supporting an extensive 131K token context length, the model achieves efficient inference via multi-expert parallel collaboration and quantization, while advanced post-training techniques including SFT, DPO, and UPO ensure optimized performance across diverse applications with specialized routing and balancing losses for superior task handling.

Text131K context

ERNIE 4.5 VL 28B A3B

A powerful multimodal Mixture-of-Experts chat model featuring 28B total parameters with 3B activated per token, delivering exceptional text and vision understanding through its innovative heterogeneous MoE structure with modality-isolated routing. Built with scaling-efficient infrastructure for high-throughput training and inference, the model leverages advanced post-training techniques including SFT, DPO, and UPO for optimized performance, while supporting an impressive 131K context length and RLVR alignment for superior cross-modal reasoning and generation capabilities.

Text131K context

ERNIE 4.5 VL 424B A47B

ERNIE-4.5-VL-424B-A47B is a multimodal Mixture-of-Experts (MoE) model from Baidu’s ERNIE 4.5 series, featuring 424B total parameters with 47B active per token. It is trained jointly on text and image data using a heterogeneous MoE architecture and modality-isolated routing to enable high-fidelity cross-modal reasoning, image understanding, and long-context generation (up to 131k tokens). Fine-tuned with techniques like SFT, DPO, UPO, and RLVR, this model supports both “thinking” and non-thinking inference modes. Designed for vision-language tasks in English and Chinese, it is optimized for efficient scaling and can operate under 4-bit/8-bit quantization.

Text131K context

ERNIE 4.5 300B A47B

ERNIE-4.5-300B-A47B is a 300B parameter Mixture-of-Experts (MoE) language model developed by Baidu as part of the ERNIE 4.5 series. It activates 47B parameters per token and supports text generation in both English and Chinese. Optimized for high-throughput inference and efficient scaling, it uses a heterogeneous MoE structure with advanced routing and quantization strategies, including FP8 and 2-bit formats. This version is fine-tuned for language-only tasks and supports reasoning, tool parameters, and extended context lengths up to 131k tokens. Suitable for general-purpose LLM applications with high reasoning and throughput demands.

Text131K context

Baidu: ERNIE 4.5 21B A3B Thinking

Activity

Frequently asked questions

Explore more models

More models from baidu

More models from baidu

Baidu: ERNIE 4.5 21B A3B Thinking

Activity

Frequently asked questions

What is ERNIE 4.5 21B A3B Thinking?

What is the context length of ERNIE 4.5 21B A3B Thinking?

What other text models does Baidu have?

When was ERNIE 4.5 21B A3B Thinking released?

Explore more models

More models from baidu

More models from baidu