Jahanzaib
Baidu: ERNIE 4.5 21B A3B

Baidu: ERNIE 4.5 21B A3B

byBaidu

ERNIE 4.5 21B A3B is a language model from Baidu released on Jun 30, 2025. It reads text and writes text, with a 120K token context window and up to 8K tokens per reply.

Show more

It costs $0.07 per million input tokens and $0.28 per million output tokens.

It supports tool calling, and its weights are open, so you can host it yourself.

Modalities
In / out price
$0.07 / $0.28 per 1M
Context
120K
Max output
8K
Released
Jun 30, 2025
Knowledge cutoff
n/a
  • Tool calling
  • Structured output
  • Reasoning
  • Prompt caching
  • Open weights
  • Free to use

ERNIE 4.5 21B A3B pricing

USD, as listed on Oct 15, 2026. Most bills are input heavy, so the cached input rate matters as much as the headline price.

Input$0.07 / 1M tokens
Output$0.28 / 1M tokens

What ERNIE 4.5 21B A3B costs per month

Pick a workload close to yours or type your own volumes.

ERNIE 4.5 21B A3B per month
$8.19
Without caching
$8.19
Per request
$0.0003

Supported request parameters

What you can set when you call ERNIE 4.5 21B A3B through an OpenAI compatible API.

  • Toolstools
  • Temperaturetemperature

Alternatives to ERNIE 4.5 21B A3B

Models that take the same inputs and call tools, closest in price, cheaper ones first.

  • Mistral: Ministral 3 3B

    Index 4.8

    Reads text and images. Tool calling, structured output. Served by 9 providers.

    by mistralaiDec 2, 2025131K context$0.10/M input tokens$0.10/M output tokens9 providers

  • Meta: Llama 3.2 3B Instruct

    Reads text. Tool calling, structured output. Served by 6 providers.

    by meta-llamaJan 1, 2024$0.0509/M input tokens$0.335/M output tokens6 providers

  • Qwen-Omni Turbo

    Reads text, images, audio and video, writes text and audio. Tool calling. Served by 2 providers.

    by qwenJan 19, 2025$0.07/M input tokens$0.27/M output tokens2 providers

  • Nex AGI: Nex-N2.5-Pro

    Reads text and images. Tool calling, structured output, reasoning.

    by nex-agiSep 8, 2026$0.075/M input tokens$0.25/M output tokens

  • IBM: Granite-4.0-H-Small

    Reads text. Tool calling, structured output.

    by ibm-graniteOct 2, 2025$0.0636/M input tokens$0.265/M output tokens

  • Qwen3.5-Flash

    Reads text, images and video. Tool calling, structured output, reasoning. Served by 11 providers.

    by qwenFeb 23, 2026$0.065/M input tokens$0.26/M output tokens11 providers

More from Baidu

  • CoBuddy

    Reads text. Tool calling, reasoning.

    by baiduMay 6, 2026$0.28/M input tokens$1.13/M output tokens

  • ERNIE 5.0

    Reads text, images and video. Tool calling, reasoning.

    by baiduJan 22, 2026$0.84/M input tokens$3.37/M output tokens

  • ERNIE 4.5 300B A47B

    Reads text. Tool calling, structured output.

    by baiduJan 1, 2026$0.28/M input tokens$1.10/M output tokens

  • ERNIE 4.5 VL 424B A47B

    Reads text and images. Tool calling, reasoning. Served by 2 providers.

    by baiduJun 30, 2025$0.42/M input tokens$1.25/M output tokens2 providers

All Baidu models

ERNIE 4.5 21B A3B questions

How much does the ERNIE 4.5 21B A3B API cost?

$0.07 per million input tokens and $0.28 per million output tokens. The estimator on this page turns that into a monthly bill for your own workload.

What is the context window of ERNIE 4.5 21B A3B?

120,000 tokens of input, and it can write up to 8,000 tokens in one reply. As a rough guide, a million tokens is around 750,000 English words.

Does ERNIE 4.5 21B A3B support tool calling?

Yes. It accepts tool definitions, which is what an agent needs to call your APIs.

Is ERNIE 4.5 21B A3B open source?

Its weights are published, so you can run it on your own hardware or a cloud of your choice. Read the license before commercial use.

When was ERNIE 4.5 21B A3B released?

Jun 30, 2025.

Catalog data from models.dev (MIT licensed), refreshed daily, last on Oct 15, 2026. The summary is written from these facts.