BaiduLanguage modelAvailable

ERNIE-Light-DT-1.0

Summary

ERNIE-Light-DT-1.0 processes text and uses Baidu's Sparse Attention method, BSA. Feedback-driven conversation-history compression is intended to reduce storage and computation. Baidu lists output up to 128K tokens and a reasoning mode that can be enabled or disabled. The product page does not publish a numeric context limit.

Qianfan gives the version name ERNIE-Light-DT-1.0 and provides API access. Its official rate is CNY 0.008 per 1,000 input tokens and CNY 0.028 per 1,000 output tokens. OpenRouter lists it as `baidu/ernie-light-dt-1.0-intl`; Baidu does not confirm the `-intl` suffix on the reviewed product page.

Similar models

More models of the same kind as ERNIE-Light-DT-1.0, chosen by model line, developer, size, price and release date.

All specifications

Architecture

Architecture: Baidu cites Baidu Sparse Attention (BSA) and feedback-driven conversation-history compression intended to reduce storage and indexing computation.; Source: Architecture

Input and output

Input: Text.; Source: Input
Output: Text.; Source: Output
Context window: The reviewed Baidu product page describes a long-context model but gives no numeric token count.; Source: Context window
Reasoning: The reasoning mode can be enabled or disabled.; Source: Reasoning

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking: Unknown; no source
Watermarking technique: Not known; no source

Licence and access

Shutdown: None announced; no source

Further details

Qianfan version name: The official pricing list gives the version name ERNIE-Light-DT-1.0. It does not specify a separate HTTP request field.; Source: Qianfan version name
OpenRouter ID: OpenRouter lists the model as baidu/ernie-light-dt-1.0-intl. The -intl suffix is not confirmed in the reviewed Baidu sources.; Source: OpenRouter ID
Maximum output: Up to 128K tokens.; Source: Maximum output

Prices

Input price: CNY 0.008 per 1,000 input tokens.Source
Cached input price: CNY 0.0008 per 1,000 cached tokens.Source
Output price: CNY 0.028 per 1,000 output tokens.Source

Sources and data date

Type	Evidence and data date
Input	Documented modalities. Retrieved November 12, 2026. Source: Baidu Qianfan model overview · Baidu Qianfan model overview (retrieved November 12, 2026) · Baidu Qianfan model overview · Editorial description reviewed November 12, 2026
Output	Documented modalities. Retrieved November 12, 2026. Source: Baidu Qianfan model overview · Baidu Qianfan model overview (retrieved November 12, 2026) · Baidu Qianfan model overview · Editorial description reviewed November 12, 2026
Additional source	Baidu Qianfan token pricing (retrieved November 12, 2026) · Baidu Qianfan token pricing · Editorial description reviewed November 12, 2026
Additional source	OpenRouter ERNIE-Light-DT-1.0 listing (retrieved November 12, 2026) · OpenRouter ERNIE-Light-DT-1.0 listing · Editorial description reviewed November 12, 2026