Navigate
Advertise here

ERNIE-Light-DT-1.0 (high) vs Qwen4 Max (low)

The average score is effectively tied at 8.6 vs 8.7. ERNIE-Light-DT-1.0 (high) has the lower benchmark cost at $1.027 vs $1.064. ERNIE-Light-DT-1.0 (high) is faster at 25.66s vs 56.68s, with pass rates of 75.4% vs 87.0%.

Last updated at: 2026-11-16

Compared models

Rank
#98
Total Output Tokens
194,474
Response Time (avg)
25.66s
Total Cost
$1.027
Rank
#94
Total Output Tokens
144,069
Response Time (avg)
56.68s
Total Cost
$1.064
Recommended model ERNIE-Light-DT-1.0 (high)

It has the best score here (8.6), while responding about 2.2x faster than Qwen4 Max (low).

Detailed comparison

Metric ERNIE-Light-DT-1.0 ERNIE-Light-DT-1.0 high Release: 2026-11-12 Qwen4 Max Qwen4 Max low Release: 2026-11-16
Score 8.6 8.7
Rank #98 #94
Reliability 9.3 9.9
Consistency 9.3 10.0
Attempts 69/69 69/69
Tests Correct
Attempt pass rate 75.4% 87.0%
Flaky tests 2 0
Total Runs 69 69
Cost per result 6.418 5.318
Total Cost $1.027 $1.064
Input Price $1.130 / 1M $1.500 / 1M
Output Price $3.941 / 1M $4.500 / 1M
Cache Read Price $0.113 / 1M $0.075 / 1M
Cache Write Price N/A $1.875 / 1M
Total Input Tokens 230,563 276,790
Output Tokens 6,576 8,266
Reasoning Tokens 187,898 135,803
Response Time (avg) 25.66s 56.68s
Response Time (max) 200.80s 564.77s
Response Time (total) 590.18s 1303.63s
Parameters ~320B total (~18B active) ~2.4T total (~95B active)
Availability Closed Closed

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

Model generation showcase

Hamster playing table tennis

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

#98 ERNIE-Light-DT-1.0

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

#94 Qwen4 Max

low
Cost
$0.064
Time
239.1s
Tokens
14,300 tok

Top Models by Score

Score vs Total Cost

Response Time (avg)

Score vs Response Time (avg)

Total Output Tokens

Score vs Total Output Tokens

Category Breakdown

Coding Score Consistency Attempt pass rate Flaky tests Tests Correct Response Time (avg) Input Tokens Output Tokens Reasoning Tokens
ERNIE-Light-DT-1.0 10.0 10.0 100.0% 0 14.96s 7,317 419 16,035
Qwen4 Max 10.0 10.0 100.0% 0 15.29s 8,001 427 8,546

Switch Comparison Pair