ERNIE 4.5 VL is part of Baidu’s ERNIE 4.5 family, a suite of multimodal models capable of interpreting complex data across multiple modalities. The VL variant specifically merges vision and language processing into a unified framework, allowing the model to analyze images, comprehend text, and generate detailed responses that connect both modalities. This model is suitable for applications ranging from interactive visual Q&A and content generation to document interpretation and visual reasoning. As an open-source solution under the Apache 2.0 license, it offers broad flexibility for research, development, and commercial applications.
ERNIE 4.5 VL combines cutting-edge multimodal reasoning with practical deployment flexibility. It offers open-source accessibility under Apache 2.0, a scalable architecture that ranges from efficient 28B models to high-capacity 424B models, and the ability to handle extremely long contexts for complex tasks. Its ecosystem includes tools and frameworks that accelerate development and integration, making it one of the most versatile solutions for vision-language AI available today.
Input: $0.5577 per 1M tokens
Output: $1.677 per 1M tokens
Measured on the AI/ML API gateway across live traffic. Figures are medians unless stated. Last updated 2026-10-12.
Image, document, and chart understanding are documented use cases for ERNIE 4.5 VL 424B A47B, including visual reasoning and visual question answering.
It is also documented for cross-modal reasoning, with the model trained on text and image data to handle tasks that combine textual and visual information.
The model is documented as suitable for reasoning-centered multimodal tasks, with strong performance on benchmarks such as MathVista, MMMU, and VisualPuzzle.
It is also documented for perception-focused visual tasks, with good results on datasets such as CV-Bench and RealWorldQA.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "baidu/ernie-4.5-vl-424b-a47b", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "baidu/ernie-4.5-vl-424b-a47b", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"baidu/ernie-4.5-vl-424b-a47b","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
ERNIE 4.5 VL 424B A47B This page | Coding + agents | |||
| Reasoning + agents | ||||
| Balanced coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Reasoning + agents |
ERNIE 4.5 VL 424B A47B has a 123,000 tokens context window and can return up to 123,000 tokens.
ERNIE 4.5 VL 424B A47B takes image, text as input and returns text.
Use baidu/ernie-4.5-vl-424b-a47b as the model id. Requests go to https://api.aimlapi.com/v1/chat/completions.
ERNIE 4.5 VL 424B A47B became available on January 12, 2026.
ERNIE 4.5 VL 424B A47B is priced at input $0.5577 / 1M tokens, output $1.677 / 1M tokens.
Yes, ERNIE 4.5 VL 424B A47B can stream responses as they are generated.
Yes, ERNIE 4.5 VL 424B A47B accepts image input alongside text.
ERNIE 4.5 VL 424B A47B was built by Baidu.
