COMPARE.

zai

GLM-4.5

Models

Context window

131,072 tokens

Maximum output

32,000 tokens^*

Open weights

Yes

Maximum input

128,000 tokens^*

Arena 1430.1Human preference · 23,746 votes

zai

GLM-4.5-Air

Models

Context window

131,072 tokens

Maximum output

32,000 tokens^*

Open weights

Yes

Maximum input

128,000 tokens^*

Arena 1383.7Human preference · 30,599 votes

zai

GLM-4.5-Air-Base

Models

Context window

131,072 tokens

Open weights

Yes

Model license

mit

Checkpoint tensor payload

205.76 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

GLM-4.5-Base

Models

Context window

131,072 tokens

Open weights

Yes

Model license

mit

Checkpoint tensor payload

667.46 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

GLM-4.5V

Models

Context window

65,536 tokens

Maximum output

32,000 tokens^*

Open weights

Yes

Maximum input

128,000 tokens^*

Arena 1332.7Human preference · 4,810 votes

zai

GLM-4.6

Models

Context window

202,752 tokens

Maximum output

128,000 tokens^*

Open weights

Yes

Maximum input

200,000 tokens^*

Arena 1440.2Human preference · 35,660 votes

zai

GLM-4.6V

Models

Context window

131,072 tokens

Open weights

Yes

Model license

mit

Checkpoint tensor payload

200.63 GiB

Arena 1376.5Human preference · 2,825 votes

zai

GLM-4.6V-Flash

Models

Context window

131,072 tokens

Open weights

Yes

Model license

mit

Checkpoint tensor payload

19.17 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

GLM-4.7-Flash

Models

Context window

202,752 tokens

Maximum output

128,000 tokens^*

Open weights

Yes

Maximum input

Reported value 1: 203,000 tokens^*

Reported value 2: 200,000 tokens^*

Arena 1350.8Human preference · 11,939 votes

zai

GLM-5.3-BF16

Models

Context window

1,048,576 tokens

Open weights

Yes

Model license

other

Checkpoint tensor payload

1,403.19 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

GLM-5.3-Flash-BF16

Models

Context window

1,048,576 tokens

Open weights

Yes

Model license

mit

Checkpoint tensor payload

598.51 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

glm-edge-1.5b-chat

Models

Context window

8,192 tokens

Open weights

Yes

Model license

other

Checkpoint tensor payload

2.97 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

glm-edge-4b-chat

Models

Context window

8,192 tokens

Open weights

Yes

Model license

other

Checkpoint tensor payload

8.06 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

glm-edge-v-2b

Models

Context window

4,096 tokens

Open weights

Yes

Model license

other

Checkpoint tensor payload

3.86 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

glm-edge-v-5b

Models

Context window

4,096 tokens

Open weights

Yes

Model license

other

Checkpoint tensor payload

9.05 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

GLM-Z1-32B-0414

Models

Context window

32,768 tokens

Open weights

Yes

Model license

mit

Checkpoint tensor payload

60.66 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

GLM-Z1-9B-0414

Models

Context window

32,768 tokens

Open weights

Yes

Model license

mit

Checkpoint tensor payload

17.51 GiB

Radar · unrated0/4 core tests · not enough evidence

zai

GLM-Z1-Rumination-32B-0414

Models

Context window

131,072 tokens

Open weights

Yes

Model license

mit

Checkpoint tensor payload

61.73 GiB

Radar · unrated0/4 core tests · not enough evidence

Unspecified

global.cohere.embed-v4:0

Models^*

Maximum input

128,000 tokens^*

API mode

Embeddings^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

global.twelvelabs.pegasus-1-5-v1:0

Models^*

Maximum output

98,304 tokens^*

Maximum input

261,120 tokens^*

API mode

Chat^*

Video input

Yes^*

Radar · unrated0/4 core tests · not enough evidence

zai

Glyph

Models

Context window

131,072 tokens

Open weights

Yes

Model license

mit

Checkpoint tensor payload

19.17 GiB

Radar · unrated0/4 core tests · not enough evidence

Unspecified

gmi/deepseek-ai/DeepSeek-V3-0324

Models^*

Maximum output

16,384 tokens^*

Maximum input

163,840 tokens^*

Tool calling

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

gmi/deepseek-ai/DeepSeek-V3.2

Models^*

Maximum output

16,384 tokens^*

Maximum input

163,840 tokens^*

Tool calling

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

gmi/openai/gpt-5

Models^*

Maximum output

32,000 tokens^*

Maximum input

409,600 tokens^*

Tool calling

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence