How do you use ERNIE 4.5 21B A3B Thinking?

ERNIE 4.5 21B A3B Thinking can be called through OpenRouter's OpenAI-compatible API using the model ID baidu/ernie-4.5-21b-a3b-thinking. This page includes a ready-to-run curl example and the supported parameter list.

Is ERNIE 4.5 21B A3B Thinking free?

ERNIE 4.5 21B A3B Thinking is a paid model in the OpenRouter catalog, with separate input and output token pricing shown on this page.

ERNIE 4.5 21B A3B Thinking

All models

BaiduQianfan

ERNIE 4.5 21B A3B Thinking

Q: What is ERNIE 4.5 21B A3B Thinking?

ERNIE 4.5 21B A3B Thinking is an AI model from Baidu in the Qianfan series. Agent Mag tracks its context window, pricing, modalities, and supported API parameters on this page.

131K context

$0.070/M input

$0.280/M output

ERNIE 4.5 21B A3B Thinking is an AI model from Baidu built for agent workflows, with support for text input and text output. 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.

What is ERNIE 4.5 21B A3B Thinking?

ERNIE 4.5 21B A3B Thinking is an AI model from Baidu that Agent Mag tracks for pricing, context window, modalities, benchmarks, and API compatibility. Builders can use this page to compare ERNIE 4.5 21B A3B Thinking against other models for agent workflows and production deployments.

Model ID

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

Input

text

Output

text

Supported Parameters

frequency_penaltyinclude_reasoningmax_tokenspresence_penaltyreasoningrepetition_penaltyseedstoptemperaturetop_ktop_p

More from Baidu

Qianfan-OCR-Fast (free)

66K ctxFree

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.

ERNIE 4.5 21B A3B

120K ctx$0.070/M

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...

ERNIE 4.5 VL 28B A3B

30K ctx$0.140/M

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....

ERNIE 4.5 300B A47B

123K ctx$0.280/M

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...

Stay in the know

ERNIE 4.5 21B A3B Thinking

What is ERNIE 4.5 21B A3B Thinking?

More from Baidu

Related content