GLM 4.5

Jul 2025
GLM 4.5
Compare in
on
—per task?
$0.60 / $2.20per 1M in / out?cache $0.11
—output speed?
131Kcontext

Rankingin thinking mode?

Strong at

B
Reasoning#155 of 334Science#124 of 312Knowledge#143 of 318
C
Tool calling#123 of 242Long context#172 of 273Instructions#135 of 245

Alternatives?

ModelScoreCostPer taskWhat changes
CheaperMiMo v2.6 ProthinkingOpen×3.6-46%$0.13Memory (4-bit): 215 GB → 615 GB+4BetterGPT-6 Lunalow+69%-80%$0.005Function calling only with none · Chat Completions+3OpenDeepSeek V4.1 Flash 1017max×3.5-48%$0.26Memory (4-bit): 215 GB → 458 GB+5CheaperClaude Haiku 5.5max×3.4-80%$0.21Remove sampling parameters+5CheaperMiniMax M3.1 Flashthinking×3.3-74%$0.22Complete model response must be appended

Run it yourself

estimate
Fits on✗ 8 GB (phone, laptop)✗ 16 GB✗ 24 GB GPU✗ 48 GB GPU✗ 80 GB GPU
LicenceMITfree for commercial use

History

10 versions
1020304050Oct 25Jan 26Apr 26Jul 26Release dateAA Intelligence Index 4.3.24.54.5V4.64.6V4.755.15.25.3
1020304050Jan 26Jul 26Release dateAA Intelligence Index 4.3.24.54.5V4.64.6V4.755.15.25.3

Benchmarks

all categories
Chat & Knowledge
Humanity's Last Exam13#156
AA Omniscience Index-27.4#144
CritPt (physics)0#180
AA Intelligence Index 4.3.2≈12.8#170
AA Omniscience · Accuracy25.1#111
AA Omniscience · Hallucination Rate70.1#134
AA-LCR (long context) 1.152.7#173
MMLU-Pro · official boardlab84.6
GPQA Diamond78.2#125
AIME 2025 · AA archived73.7#59
Coding
Terminal-Bench Hard (version unstated)22#86
LiveCodeBench · AA archived73.8#33
Agents & Tool Use
τ²-bench · AA43#125
Writing & Language
IFBench (instructions)44.1#136

FAQ

How much does GLM 4.5 cost?›

$0.60 per 1M input tokens and $2.20 per 1M output tokens; cached input $0.11 per 1M; cheapest via DeepInfra at $0.38 / $1.60.

Is GLM 4.5 free to use?›

There is no free API; the cheapest paid price is $0.38 per 1M input tokens. The weights are open, so you can run it yourself.

How good is GLM 4.5, and what is it best at?›

#155 of 334 in Chat & Knowledge, #86 of 239 in Coding, #123 of 242 in Agents & Tool Use, #87 of 195 in Writing & Language.

Is GLM 4.5 better than Nemotron 3 Super 120B A12B?›

On AA Intelligence Index 4.3.2, GLM 4.5 scores 12.8 vs 12.8 — about the same.

How can I use GLM 4.5?›

Through the API of DeepInfra.

What is the context window of GLM 4.5?›

131K tokens, with up to 98K output tokens.

When was GLM 4.5 released? Is there a newer version?›

Released 2025-07-28. It is the current version in its line.

Is GLM 4.5 open source?›

Yes, the weights are open (MIT licence).

Can I run GLM 4.5 locally?›

The weights are open and can be self-hosted; hardware needs are not confirmed yet.

Can GLM 4.5 read images?›

No, it accepts text only.

What is GLM 4.5?›

GLM 4.5 is an open-weights language model from Z.ai, released Jul 2025.

What are the best alternatives to GLM 4.5?›

MiMo v2.6 Pro (cheaper), GPT-6 Luna (better), DeepSeek V4.1 Flash 1017 (open weights).