FlagshipBaidu

ERNIE 4.5 300B

ERNIE 4.5 300B is Baidu's flagship large language model with 300 billion parameters and a 123K token context window for advanced text generation.

Context 123K
Tier Flagship
Input from
$0.280 / 1M tokens
across 1 provider
Compare Prices

API Pricing

ProviderInput / 1MOutput / 1MSpeedTTFTUpdated
OpenRouter logo
OpenRouter
$0.280$1.1024.2 t/s1.8s4/14/2026

Prices updated daily. Last check: 4/14/2026

Model Details

General

Creator
Baidu
Family
ERNIE
Tier
Flagship
Context Window
123K
Modalities
Text

Capabilities

Tool Calling
No
Open Source
No

Strengths & Limitations

  • 300 billion parameters provide extensive model capacity for complex tasks
  • 123K token context window supports lengthy document processing
  • 25.62 tokens per second output speed for reasonable inference performance
  • Flagship tier positioning with advanced reasoning capabilities
  • Strong Chinese language support from Baidu's expertise
  • Large parameter count enables sophisticated text generation
  • Extended context length allows for comprehensive document analysis
  • No tool calling or function calling capabilities
  • Text-only modality without image or multimodal input support
  • Proprietary model with no open source availability
  • 1,537ms time to first token is slower than some competing models
  • Limited ecosystem compared to OpenAI or Anthropic models

Key Features

•300 billion parameter architecture
•123K token context window
•Text input and output processing
•Multilingual language support with Chinese specialization
•Streaming response capability
•Large-scale reasoning and generation
•Extended context document processing
•Flagship-tier language understanding

About ERNIE 4.5 300B

ERNIE 4.5 300B is Baidu's flagship large language model in the ERNIE family, featuring 300 billion parameters for handling complex text generation and reasoning tasks. The model represents Baidu's most capable offering in their ERNIE model lineup, designed to compete with other flagship models in the market. The model supports a 123K token context window for processing lengthy documents and maintaining extended conversations. ERNIE 4.5 300B focuses exclusively on text modalities and delivers benchmark performance of 25.62 output tokens per second with a time to first token of 1,537 milliseconds. The model is particularly oriented toward Chinese language tasks given Baidu's focus on the Chinese market, though it supports multilingual capabilities. ERNIE 4.5 300B is positioned for applications requiring sophisticated language understanding and generation, competing in the flagship tier alongside other large-scale models. Organizations typically deploy this model for complex reasoning, content creation, and applications where the large parameter count and substantial context window provide advantages over smaller alternatives.

Common Use Cases

ERNIE 4.5 300B is designed for organizations requiring sophisticated text processing and generation capabilities, particularly those working with Chinese language content or needing extensive context windows. The model's 300 billion parameters and 123K context length make it suitable for complex reasoning tasks, long-form content creation, document analysis, and applications requiring deep language understanding. Its flagship positioning targets use cases where model sophistication is prioritized over speed, such as research applications, advanced content generation, comprehensive document summarization, and complex question-answering systems that benefit from the extended context window and large parameter count.

Frequently Asked Questions

How much does ERNIE 4.5 300B cost per million tokens?

ERNIE 4.5 300B pricing varies by provider and deployment method. Check the pricing table above for current rates across available providers offering this model.

What is ERNIE 4.5 300B best used for?

ERNIE 4.5 300B excels at complex reasoning tasks, long-form content generation, and document analysis requiring its 123K context window. The 300 billion parameter architecture makes it particularly effective for sophisticated text processing, Chinese language applications, and use cases where model capability is more important than response speed.

Does ERNIE 4.5 300B support function calling or multimodal input?

No, ERNIE 4.5 300B does not support function calling or tool use capabilities. It also operates exclusively with text modalities and cannot process images or other non-text inputs, focusing purely on advanced text generation and reasoning tasks.