Advertise here
#97

ERNIE-Light-DT-1.0

Baidu Release: 2026-11-12 Tested on: 2026-11-12 23:17 baidu/ernie-light-dt-1.0-intl::high
~320B total (~18B active)MoEClosedEstimated
(max) (high) (low) (none)

Summary

ERNIE-Light-DT-1.0 scores 8.6 on AI BENCHY and ranks #97. It has 9.3 reliability, a 75.4% pass rate, $1.027 total cost, and 25.66s average response time.

Model facts

Researched on 2026-11-12

Estimated
Parameters
~320B total (~18B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. OpenRouter states the model is based on open-source GLM models with Baidu Sparse Attention but names no backbone version or size; no Baidu model card or weights were found. The 320B total / 18B active MoE values assume a GLM-5.3-Flash backbone and are low confidence. API-only exact checkpoint. OpenRouter declares optional reasoning with none, low, high and max efforts, max default.

Parameter source 1 Parameter source 2 Availability source

Score

8.6

Consistency

9.3

Reliability

9.3

Total Cost (Current Price)

$1.027

Total Output Tokens

194,474

Total Input Tokens

230,563

Input Price

$1.130 / 1M

Output Price

$3.941 / 1M

Cache Read Price

$0.113 / 1M

Cache Write Price

N/A

Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.

Tests Correct

Wrong Tests: 7

Attempt pass rate: 75.4%

Flaky tests

2

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

25.66s

Response Time (max): 200.80s

Response Time (total): 590.18s

Wrong answer: 5 Extra formatting: 1 No answer: 1

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#97 ERNIE-Light-DT-1.0

high
Cost
$0.067
Time
129.8s
Tokens
16,960 tok

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

ERNIE-Light-DT-1.0highvsMuse EmbermaxERNIE-Light-DT-1.0highvsDeepSeek V4.1 FlashmaxERNIE-Light-DT-1.0highvsMuse EmberlowERNIE-Light-DT-1.0highvsGPT-6.1 TerranoneERNIE-Light-DT-1.0highvsClaude Fable 5.2highERNIE-Light-DT-1.0highvsMuse Spark 1.1lowERNIE-Light-DT-1.0highvsGPT-6 SollowERNIE-Light-DT-1.0highvsClaude Fable 5.2lowERNIE-Light-DT-1.0highvsSeed 2.1 TurbomediumERNIE-Light-DT-1.0highvsClaude Sonnet 5.5high

Category Breakdown

Category Score Consistency Tests Correct
Agentic 10.0 10.0
Anti-AI Tricks 10.0 10.0
Coding 10.0 10.0
Combined 6.5 10.0
Data parsing and extraction 10.0 10.0
Domain specific 3.6 7.2
General Intelligence 6.5 3.4
Instructions following 10.0 10.0
Puzzle Solving 7.7 10.0
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models