GLM-4.5
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
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
1,464.74 rating
Confidence interval: 1,450.36 – 1,479.13
1785 votes
- Category
- chinese
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
GLM-4.5
1,433.89 rating
Confidence interval: 1,425.14 – 1,442.64
4741 votes
- Category
- coding
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
GLM-4.5
1,395.00 rating
Confidence interval: 1,384.24 – 1,405.77
3096 votes
- Category
- creative_writing
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
GLM-4.5
1,436.54 rating
Confidence interval: 1,429.99 – 1,443.08
10120 votes
- Category
- english
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
GLM-4.5
1,422.47 rating
Confidence interval: 1,415.70 – 1,429.24
16956 votes
- Category
- exclude_ties
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
GLM-4.5
1,433.15 rating
Confidence interval: 1,415.82 – 1,450.49
1157 votes
- Category
- expert
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating
GLM-4.5
1,418.26 rating
Confidence interval: 1,386.43 – 1,450.08
341 votes
- Category
- french
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating
GLM-4.5
1,406.77 rating
Confidence interval: 1,381.39 – 1,432.15
539 votes
- Category
- german
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
GLM-4.5
1,429.19 rating
Confidence interval: 1,422.83 – 1,435.55
11112 votes
- Category
- hard_prompts
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
GLM-4.5
1,429.39 rating
Confidence interval: 1,420.67 – 1,438.11
4831 votes
- Category
- hard_prompts_english
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
GLM-4.5
1,419.93 rating
Confidence interval: 1,410.62 – 1,429.25
4225 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating
GLM-4.5
1,404.60 rating
Confidence interval: 1,395.52 – 1,413.67
4321 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating
GLM-4.5
1,437.96 rating
Confidence interval: 1,422.27 – 1,453.64
1476 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating
GLM-4.5
1,444.11 rating
Confidence interval: 1,434.43 – 1,453.79
3859 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating
GLM-4.5
1,429.98 rating
Confidence interval: 1,413.13 – 1,446.82
1260 votes
- Category
- industry_mathematical
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating
GLM-4.5
1,454.73 rating
Confidence interval: 1,438.41 – 1,471.04
1386 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating
GLM-4.5
1,440.29 rating
Confidence interval: 1,433.19 – 1,447.39
8089 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating
GLM-4.5
1,402.60 rating
Confidence interval: 1,394.23 – 1,410.96
5304 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating
GLM-4.5
1,404.14 rating
Confidence interval: 1,396.28 – 1,412.01
6164 votes
- Category
- instruction_following
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating
GLM-4.5
1,415.20 rating
Confidence interval: 1,392.52 – 1,437.89
776 votes
- Category
- japanese
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating
GLM-4.5
1,379.85 rating
Confidence interval: 1,355.97 – 1,403.74
655 votes
- Category
- korean
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating
GLM-4.5
1,412.45 rating
Confidence interval: 1,403.93 – 1,420.98
5087 votes
- Category
- longer_query
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating
GLM-4.5
1,427.21 rating
Confidence interval: 1,411.58 – 1,442.84
1402 votes
- Category
- math
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating
GLM-4.5
1,414.90 rating
Confidence interval: 1,405.02 – 1,424.78
3704 votes
- Category
- multi_turn
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating
GLM-4.5
1,416.99 rating
Confidence interval: 1,411.08 – 1,422.90
13665 votes
- Category
- non_english
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating
GLM-4.5
1,430.00 rating
Confidence interval: 1,425.11 – 1,434.89
23797 votes
- Category
- overall
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating
GLM-4.5
1,422.43 rating
Confidence interval: 1,407.58 – 1,437.27
1676 votes
- Category
- polish
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating
GLM-4.5
1,412.93 rating
Confidence interval: 1,398.68 – 1,427.18
1568 votes
- Category
- russian
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating
GLM-4.5
1,454.29 rating
Confidence interval: 1,430.98 – 1,477.61
678 votes
- Category
- spanish
- Evaluated configuration
- GLM-4.5
- Evaluation type
- human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating
GLM-4.5
1,464.83 rating
Confidence interval: 1,450.43 – 1,479.22
1783 votes
- Category
- chinese
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
GLM-4.5
1,434.07 rating
Confidence interval: 1,425.32 – 1,442.82
4742 votes
- Category
- coding
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
GLM-4.5
1,394.91 rating
Confidence interval: 1,384.14 – 1,405.69
3093 votes
- Category
- creative_writing
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
GLM-4.5
1,436.52 rating
Confidence interval: 1,429.98 – 1,443.07
10120 votes
- Category
- english
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
GLM-4.5
1,422.45 rating
Confidence interval: 1,415.69 – 1,429.22
16953 votes
- Category
- exclude_ties
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-15 · rating
GLM-4.5
1,433.19 rating
Confidence interval: 1,415.86 – 1,450.52
1157 votes
- Category
- expert
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating
GLM-4.5
1,418.96 rating
Confidence interval: 1,387.10 – 1,450.81
340 votes
- Category
- french
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating
GLM-4.5
1,406.73 rating
Confidence interval: 1,381.35 – 1,432.10
539 votes
- Category
- german
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
GLM-4.5
1,429.12 rating
Confidence interval: 1,422.76 – 1,435.48
11110 votes
- Category
- hard_prompts
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
GLM-4.5
1,429.28 rating
Confidence interval: 1,420.56 – 1,438.01
4832 votes
- Category
- hard_prompts_english
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating
GLM-4.5
1,420.00 rating
Confidence interval: 1,410.69 – 1,429.32
4226 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
GLM-4.5
1,404.37 rating
Confidence interval: 1,395.29 – 1,413.45
4315 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
GLM-4.5
1,437.59 rating
Confidence interval: 1,421.90 – 1,453.28
1475 votes
- Category
- industry_legal_and_government
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
GLM-4.5
1,444.11 rating
Confidence interval: 1,434.43 – 1,453.79
3858 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
GLM-4.5
1,430.02 rating
Confidence interval: 1,413.18 – 1,446.87
1260 votes
- Category
- industry_mathematical
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
GLM-4.5
1,454.69 rating
Confidence interval: 1,438.37 – 1,471.01
1385 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
GLM-4.5
1,440.42 rating
Confidence interval: 1,433.33 – 1,447.52
8092 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
GLM-4.5
1,402.46 rating
Confidence interval: 1,394.09 – 1,410.83
5301 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
GLM-4.5
1,404.01 rating
Confidence interval: 1,396.15 – 1,411.88
6162 votes
- Category
- instruction_following
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating
GLM-4.5
1,415.44 rating
Confidence interval: 1,392.75 – 1,438.13
775 votes
- Category
- japanese
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating
GLM-4.5
1,380.40 rating
Confidence interval: 1,356.52 – 1,404.27
656 votes
- Category
- korean
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
GLM-4.5
1,412.45 rating
Confidence interval: 1,403.92 – 1,420.98
5085 votes
- Category
- longer_query
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
GLM-4.5
1,427.24 rating
Confidence interval: 1,411.61 – 1,442.87
1402 votes
- Category
- math
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
GLM-4.5
1,414.96 rating
Confidence interval: 1,405.08 – 1,424.84
3706 votes
- Category
- multi_turn
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
GLM-4.5
1,416.99 rating
Confidence interval: 1,411.08 – 1,422.90
13664 votes
- Category
- non_english
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
GLM-4.5
1,429.94 rating
Confidence interval: 1,425.05 – 1,434.83
23796 votes
- Category
- overall
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating
GLM-4.5
1,422.48 rating
Confidence interval: 1,407.64 – 1,437.32
1676 votes
- Category
- polish
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
GLM-4.5
1,412.82 rating
Confidence interval: 1,398.58 – 1,427.05
1570 votes
- Category
- russian
- Evaluated configuration
- glm-4.5
- Evaluation type
- human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating
GLM-4.5
1,454.19 rating
Confidence interval: 1,430.87 – 1,477.51
678 votes
- Category
- spanish
- Evaluated configuration
- glm-4.5
- 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
- 667.46 GiB
Input modalities
- text
Output modalities
- text
Processing & platform
- Architecture
- glm4_moe
System configuration
Detailed specifications
Specifications · 8 fields
- input cost per token
- 6.0e-7*
Scope and conditions
- native_context
- provider
- zai*
- mode
- chat*
- full_native_model_id
- zai/glm-4.5*
- 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*
- 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*
- 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*
- 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*
- 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
- 2.2e-6*
Scope and conditions
- native_context
- provider
- zai*
- mode
- chat*
- full_native_model_id
- zai/glm-4.5*
- 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*
- 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*
- 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