GLM-4.5-Air
Stored tensors can include buffers and repeated shared weights. A verified unique parameter count is unavailable. Checkpoint size is not the memory required to run the model.
GeekRadar / The comparison desk
COMPARE.
AI modelsGLM-4.5-Air
AI model
- API mode
- Chat*
- Context window
- 131,072 tokens
- Maximum output
- 32,000 tokens*
- Open weights
- Yes
Capability & preference
Human preference · Arena
Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.
Recorded results
Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.
Arena text preference / chinesehuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,425.98 rating
Confidence interval: 1,413.42 – 1,438.54
2234 votes
- Category
- chinese
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,396.75 rating
Confidence interval: 1,389.04 – 1,404.45
6099 votes
- Category
- coding
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,343.12 rating
Confidence interval: 1,333.66 – 1,352.59
3996 votes
- Category
- creative_writing
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,397.40 rating
Confidence interval: 1,391.72 – 1,403.09
13217 votes
- Category
- english
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,357.51 rating
Confidence interval: 1,351.63 – 1,363.40
21527 votes
- Category
- exclude_ties
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,370.08 rating
Confidence interval: 1,354.79 – 1,385.37
1472 votes
- Category
- expert
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,399.65 rating
Confidence interval: 1,372.52 – 1,426.77
520 votes
- Category
- french
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,377.44 rating
Confidence interval: 1,355.71 – 1,399.17
744 votes
- Category
- german
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,378.64 rating
Confidence interval: 1,373.17 – 1,384.11
14890 votes
- Category
- hard_prompts
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,387.76 rating
Confidence interval: 1,380.34 – 1,395.18
6578 votes
- Category
- hard_prompts_english
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,370.48 rating
Confidence interval: 1,362.34 – 1,378.62
5501 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,339.50 rating
Confidence interval: 1,331.56 – 1,347.44
5692 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,381.95 rating
Confidence interval: 1,368.53 – 1,395.38
1907 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,406.86 rating
Confidence interval: 1,398.37 – 1,415.34
4917 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,395.53 rating
Confidence interval: 1,380.28 – 1,410.78
1463 votes
- Category
- industry_mathematical
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,401.34 rating
Confidence interval: 1,386.95 – 1,415.73
1646 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,401.00 rating
Confidence interval: 1,394.77 – 1,407.22
10456 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,354.12 rating
Confidence interval: 1,346.78 – 1,361.47
6983 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,353.74 rating
Confidence interval: 1,346.91 – 1,360.58
8198 votes
- Category
- instruction_following
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,347.70 rating
Confidence interval: 1,326.94 – 1,368.47
845 votes
- Category
- japanese
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,307.86 rating
Confidence interval: 1,284.71 – 1,331.02
673 votes
- Category
- korean
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,365.93 rating
Confidence interval: 1,358.47 – 1,373.39
6767 votes
- Category
- longer_query
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,395.53 rating
Confidence interval: 1,380.74 – 1,410.33
1529 votes
- Category
- math
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,370.99 rating
Confidence interval: 1,362.47 – 1,379.51
4994 votes
- Category
- multi_turn
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,366.04 rating
Confidence interval: 1,360.84 – 1,371.25
17393 votes
- Category
- non_english
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,383.69 rating
Confidence interval: 1,379.46 – 1,387.92
30652 votes
- Category
- overall
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,367.44 rating
Confidence interval: 1,353.25 – 1,381.63
1861 votes
- Category
- polish
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,373.02 rating
Confidence interval: 1,360.39 – 1,385.64
2003 votes
- Category
- russian
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating
GLM-4.5-Air
1,385.56 rating
Confidence interval: 1,364.59 – 1,406.52
824 votes
- Category
- spanish
- Evaluated configuration
- GLM-4.5-Air
- Evaluation type
- human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,425.96 rating
Confidence interval: 1,413.40 – 1,438.52
2235 votes
- Category
- chinese
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,397.09 rating
Confidence interval: 1,389.39 – 1,404.80
6101 votes
- Category
- coding
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,343.29 rating
Confidence interval: 1,333.83 – 1,352.75
3999 votes
- Category
- creative_writing
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,397.43 rating
Confidence interval: 1,391.75 – 1,403.12
13218 votes
- Category
- english
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,357.53 rating
Confidence interval: 1,351.65 – 1,363.42
21535 votes
- Category
- exclude_ties
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,370.33 rating
Confidence interval: 1,355.04 – 1,385.63
1471 votes
- Category
- expert
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,399.17 rating
Confidence interval: 1,372.04 – 1,426.31
520 votes
- Category
- french
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,377.04 rating
Confidence interval: 1,355.30 – 1,398.78
743 votes
- Category
- german
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,378.69 rating
Confidence interval: 1,373.22 – 1,384.16
14892 votes
- Category
- hard_prompts
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,387.68 rating
Confidence interval: 1,380.26 – 1,395.11
6577 votes
- Category
- hard_prompts_english
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,370.42 rating
Confidence interval: 1,362.28 – 1,378.56
5503 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,339.50 rating
Confidence interval: 1,331.56 – 1,347.43
5697 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,381.41 rating
Confidence interval: 1,367.99 – 1,394.83
1909 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,406.83 rating
Confidence interval: 1,398.35 – 1,415.32
4921 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,395.54 rating
Confidence interval: 1,380.30 – 1,410.78
1464 votes
- Category
- industry_mathematical
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,401.44 rating
Confidence interval: 1,387.04 – 1,415.83
1645 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,401.15 rating
Confidence interval: 1,394.92 – 1,407.38
10457 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,354.18 rating
Confidence interval: 1,346.84 – 1,361.52
6987 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,353.66 rating
Confidence interval: 1,346.83 – 1,360.50
8201 votes
- Category
- instruction_following
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,347.97 rating
Confidence interval: 1,327.19 – 1,368.74
845 votes
- Category
- japanese
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,307.50 rating
Confidence interval: 1,284.36 – 1,330.65
674 votes
- Category
- korean
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,366.11 rating
Confidence interval: 1,358.65 – 1,373.57
6773 votes
- Category
- longer_query
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,395.54 rating
Confidence interval: 1,380.74 – 1,410.33
1529 votes
- Category
- math
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,371.07 rating
Confidence interval: 1,362.54 – 1,379.59
4994 votes
- Category
- multi_turn
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,366.04 rating
Confidence interval: 1,360.84 – 1,371.25
17400 votes
- Category
- non_english
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,383.69 rating
Confidence interval: 1,379.47 – 1,387.92
30660 votes
- Category
- overall
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,367.72 rating
Confidence interval: 1,353.54 – 1,381.91
1861 votes
- Category
- polish
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,372.91 rating
Confidence interval: 1,360.29 – 1,385.53
2005 votes
- Category
- russian
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating
GLM-4.5-Air
1,385.45 rating
Confidence interval: 1,364.47 – 1,406.42
824 votes
- Category
- spanish
- Evaluated configuration
- glm-4.5-air
- Evaluation type
- human preference
Run it locally
Capabilities
- Context window
- 131,072 tokens
- Maximum input
- 128,000 tokens*
- Maximum output
- 32,000 tokens*
- Tool calling
- Yes*
- Open weights
- Yes
- API mode
- Chat*
- Tool choice
- Yes*
- Model license
- mit
- Checkpoint tensor payload
- 205.76 GiB
Input modalities
- text
Output modalities
- text
Processing & platform
- Architecture
- glm4_moe
System configuration
Detailed specifications
Specifications · 8 fields
- input cost per token
- 2.0e-7*
Scope and conditions
- native_context
- provider
- zai*
- mode
- chat*
- full_native_model_id
- zai/glm-4.5-air*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- litellm provider
- zai*
Scope and conditions
- native_context
- provider
- zai*
- mode
- chat*
- full_native_model_id
- zai/glm-4.5-air*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- max input tokens
- 128,000*
Scope and conditions
- native_context
- provider
- zai*
- mode
- chat*
- full_native_model_id
- zai/glm-4.5-air*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- max output tokens
- 32,000*
Scope and conditions
- native_context
- provider
- zai*
- mode
- chat*
- full_native_model_id
- zai/glm-4.5-air*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- mode
- chat*
Scope and conditions
- native_context
- provider
- zai*
- mode
- chat*
- full_native_model_id
- zai/glm-4.5-air*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- output cost per token
- 1.1e-6*
Scope and conditions
- native_context
- provider
- zai*
- mode
- chat*
- full_native_model_id
- zai/glm-4.5-air*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- supports function calling
- Yes*
Scope and conditions
- native_context
- provider
- zai*
- mode
- chat*
- full_native_model_id
- zai/glm-4.5-air*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- supports tool choice
- Yes*
Scope and conditions
- native_context
- provider
- zai*
- mode
- chat*
- full_native_model_id
- zai/glm-4.5-air*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
0 selected Compare 0
* Provisional data