ERNIE 4.5 21B A3B Thinking vs GLM 4.6

Compare ERNIE 4.5 21B A3B Thinking from baidu and GLM 4.6 from Z.ai on key metrics including benchmarks, price, context length, and other model features. Access both models and hundreds of others through the OpenRouter API.

Favicon for baidu
ERNIE 4.5 21B A3B Thinking
Favicon for z-ai
GLM 4.6

Overview

ERNIE 4.5 21B A3B Thinking
Author
baidu
GLM 4.6
Author
Z.ai
ERNIE 4.5 21B A3B Thinking
Context length
131Ktokens
GLM 4.6
Context length
205Ktokens
ERNIE 4.5 21B A3B Thinking
Reasoning
GLM 4.6
Reasoning
ERNIE 4.5 21B A3B Thinking
Input modalities
GLM 4.6
Input modalities
ERNIE 4.5 21B A3B Thinking
Output modalities
GLM 4.6
Output modalities

Analysis

ERNIE 4.5 21B A3B Thinking vs GLM 4.6 at a glance

Index and benchmark scores are from Artificial Analysis. Each score is the model's best published result across the configurations Artificial Analysis evaluated.

ERNIE 4.5 21B A3B Thinking
Input price
Pricing varies by provider
GLM 4.6
Input price
$0.43/M tokens
ERNIE 4.5 21B A3B Thinking
Output price
Pricing varies by provider
GLM 4.6
Output price
$1.75/M tokens
ERNIE 4.5 21B A3B Thinking
Context window
131,072 tokens
GLM 4.6
Context window
204,800 tokens
ERNIE 4.5 21B A3B Thinking
Openness Index Artificial Analysis, best published result
Not available
GLM 4.6
Openness Index Artificial Analysis, best published result
44.4
ERNIE 4.5 21B A3B Thinking
Multilingual Index Artificial Analysis, best published result
Not available
GLM 4.6
Multilingual Index Artificial Analysis, best published result
85.6%
ERNIE 4.5 21B A3B Thinking
Latency (p50)
Not available
GLM 4.6
Latency (p50)
2.30 s
ERNIE 4.5 21B A3B Thinking
Throughput (p50)
Not available
GLM 4.6
Throughput (p50)
19.0 tok/s
ERNIE 4.5 21B A3B Thinking
Provided by
baidu
GLM 4.6
Provided by
Z.ai

Benchmark data updated

Frequently asked questions

Which has the longer context window, ERNIE 4.5 21B A3B Thinking or GLM 4.6?

GLM 4.6 has the longer context window at 204,800 tokens, compared with 131,072 tokens for ERNIE 4.5 21B A3B Thinking.