Baidu: ERNIE-Light-DT-1.0

Text input Text output
Author's Description

ERNIE-Light-DT-1.0 is based on open-source GLM models and includes research findings from Baidu Search's Lightning large model. It features Baidu Sparse Attention (BSA), a unique sparse attention mechanism developed by...

Key Specifications
Cost
$$$$$
Context
1M
Released
Nov 10, 2026
Speed
★
Ability
★★★★
Reliability
★★
Supported Parameters

This model supports the following parameters:

Tools Top P Temperature Response Format Tool Choice Include Reasoning Structured Outputs Reasoning Max Tokens
Features

This model supports the following features:

Structured Outputs Reasoning Tools Response Format
Performance Summary

ERNIE-Light-DT-1.0 by Baidu delivers balanced overall performance characterized by standout analytical capabilities and an extensive 1,024,000-token context window. Across evaluations, the model displays moderate speed performance, ranking in the 33rd percentile, while offering moderate pricing in the 21st percentile. Operationally, it demonstrates exceptional reliability, achieving a 98% success rate with minimal technical failures across all benchmarks. The model’s core strength lies in complex reasoning, where it achieved a perfect 100% accuracy, making it the most accurate model at its price point and speed bracket. It also posts competitive results in software development, securing a 93.0% accuracy in coding, and demonstrates dependable multi-step precision with a 76.0% accuracy in instruction following (70th percentile). In mathematics, ERNIE-Light-DT-1.0 attained a solid 92.0% accuracy, though this placed it around the median (49th percentile). A notable weakness is its latency on intensive tasks; duration was significantly higher in mathematics and coding evaluations, placing it in the lower speed percentiles for those categories. Overall, ERNIE-Light-DT-1.0 is an exceptionally dependable model best suited for complex reasoning and structured programming tasks where analytical precision outweighs raw generation speed.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $1.13
Completion $3.94
Input Cache Read $0.113

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Baidu
Baidu | baidu/ernie-light-dt-1.0-intl-20261110 1M $1.13 / 1M tokens $3.94 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by baidu