GLM 5
GLM 5Alternatives?
CheaperGemini 3 Flash Previewunstatedwins 53%-27%$0.50 / $3
BetterMuse Spark 1.2xhighwins 57%+29%$0.97Only auto tool_choice is supported+1OpenGLM 5.3maxwins 56%+39%$2.01Reasoning cannot be disabled+3CheaperGLM 5.3 FlashunstatedOpenwins 52%-85%$0.15 / $0.50Defaults to max · Self-hosted+4
CheaperMiMo v2.5 ProunstatedOpensame-65%$0.43 / $0.87Memory (4-bit): 452 GB → 614 GBGemini 3 Flash Previewunstatedwins 53%-27%$0.50 / $3GLM 5.3 FlashunstatedOpenwins 52%-85%$0.15 / $0.50Defaults to max · Self-hosted+4MiMo v2.5 ProunstatedOpensame-65%$0.43 / $0.87Memory (4-bit): 452 GB → 614 GBMiMo v2.6 FlashunstatedOpensame-89%$0.14 / $0.28Pass back reasoning_content+3
GPT-5.6 Lunaxhighsame-71%$0.085Muse Spark 1.2xhighwins 57%+29%$0.97Only auto tool_choice is supported+1GPT-5.6 Solxhighwins 53%+416%$1.18Gemini 3.8 Flashhighwins 56%-3%*$1.24Sampling parameters are ignored+1GLM 5.2maxOpenwins 53%+39%$1.47GPT-5.5highwins 55%+626%$1.54Prompts with >272K input tokensGLM 5.3maxOpenwins 56%+39%$2.01Reasoning cannot be disabled+3GLM 5.2maxOpenwins 53%+39%$1.47GLM 5.3 FlashunstatedOpenwins 52%-85%$0.15 / $0.50Defaults to max · Self-hosted+4MiMo v2.5 ProunstatedOpensame-65%$0.43 / $0.87Memory (4-bit): 452 GB → 614 GB
Kimi K2.6unstatedOpensame+10%$0.95 / $4Memory (4-bit): 452 GB → 616 GBGLM 5.3maxOpenwins 56%—$2.01Reasoning cannot be disabled+3GLM 5.2maxOpenwins 53%—$1.47GLM 5.3 FlashunstatedOpenwins 52%—$0.15 / $0.50Defaults to max · Self-hosted+4GLM 5.1unstatedOpensame—$1.40 / $4.40GLM 4.7unstatedOpenwins 47%—$0.60 / $2.20GLM 5V Turbounstatedwins 46%—$1.20 / $4GLM 4.5unstatedOpenwins 44%—$0.60 / $2.20Context: 203K → 131KGLM 4.5 AirunstatedOpenwins 34%—$0.20 / $1.10Context: 203K → 131K+1All Z.ai models →Run it yourself
estimateHistory
10 versionsBenchmarks
in Chat & KnowledgeFAQ
How much does GLM 5 cost?›
$1 per 1M input tokens and $3.20 per 1M output tokens; cached input $0.20 per 1M; cheapest via DeepInfra at $0.60 / $2.08.
Is GLM 5 free to use?›
There is no free API; the cheapest paid price is $0.60 per 1M input tokens. The weights are open, so you can run it yourself.
How good is GLM 5, and what is it best at?›
#84 of 334 in Chat & Knowledge, #39 of 44 in Coding, #8 of 242 in Agents & Tool Use, #28 of 195 in Writing & Language. Strongest at: Tool calling, Knowledge, Instructions.
Is GLM 5 better than GLM 4.7?›
On Arena Text · Korean, GLM 5 scores 1,422 vs 1,399 — higher.
How can I use GLM 5?›
Through the API of AWS Bedrock, Amazon Bedrock, Baidu, DeepInfra, Novita and 2 more.
What is the context window of GLM 5?›
203K tokens, with up to 131K output tokens.
When was GLM 5 released? Is there a newer version?›
Released 2026-02-12. It has been superseded by GLM 5.1.
Is GLM 5 open source?›
Yes, the weights are open (MIT licence).
Can I run GLM 5 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Can GLM 5 read images?›
No, it accepts text only.
What is GLM 5?›
GLM 5 is an open-weights language model from Z.ai, released Feb 2026, the successor to GLM 4.7.
What are the best alternatives to GLM 5?›
Gemini 3 Flash Preview (cheaper), Muse Spark 1.2 (better), GLM 5.3 (open weights).