geekradar
Int
Explore GeekRadar

More to explore

Good gear All topics About us Say hello
GeekRadar / The comparison desk
COMPARE.
AI models
zai

GLM-5.1

AI model Released

API mode
Chat*
Context window
202,752 tokens
Maximum output
128,000 tokens*
Open weights
Yes

Capability & preference

General capability · ECI

A cross-benchmark capability estimate. Index points, not a percentage; intervals show uncertainty.

GLM-5.1 149.9* GLM-5.1 · 95% interval 148.1–151.7

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

GLM-5.1 1462.8 GLM-5.1 · 95% interval 1459.3–1466.3 · 66,771 votes · 2026-10-28

Model performance

Radar task score: four tests, equal weight. An experimental summary of these tasks, not a universal intelligence rating.

GLM-5.1

Radar task score

Unrated

3/4 core tests · not enough evidence

4 benchmark results for this configuration

Scientific reasoning

GPQA Diamond · 25%

Expert-written science questions; strict multiple-choice scoring. Chance accuracy is 25%.

GLM-5.1 89.9*% Approx. 95% interval: 85.7–94.1%

Novel problem solving

Mystery Game Puzzles · 25%

Best-next-move puzzles in an undisclosed game variant. A narrow reasoning test, not general intelligence.

GLM-5.1 Not available

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

Private v2 mathematics problems with Python tools. Earlier versions and Tier 4 are excluded.

GLM-5.1 36.8*% Approx. 95% interval: 31.2–42.5%

Factual answers

SimpleQA Verified · 25%

Correct factual answers under a guessing prompt, graded by a model. Not a hallucination or abstention score.

GLM-5.1 34.0*% Approx. 95% interval: 31.1–36.9%

Competition mathematics

OTIS Mock AIME 2024-2025 · Separate result

Competition problems with model-extracted answers checked against an exact key. Earlier grading procedures are excluded.

GLM-5.1 93.3*% Approx. 95% interval: 86.0–100.0%

Bars show reported task scores. Whiskers are approximate 95% intervals where standard errors are available. Different reasoning budgets are not equal-cost tests. Missing results are never zero.

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-5.1

1,514.49 rating

Confidence interval: 1,505.21 – 1,523.78

4675 votes

Category
chinese
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

GLM-5.1

1,485.92 rating

Confidence interval: 1,480.55 – 1,491.29

18059 votes

Category
coding
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

GLM-5.1

1,452.53 rating

Confidence interval: 1,446.29 – 1,458.78

13020 votes

Category
creative_writing
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

GLM-5.1

1,475.48 rating

Confidence interval: 1,470.95 – 1,480.01

28384 votes

Category
english
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

GLM-5.1

1,465.43 rating

Confidence interval: 1,460.88 – 1,469.97

49773 votes

Category
exclude_ties
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

GLM-5.1

1,478.22 rating

Confidence interval: 1,470.56 – 1,485.87

7111 votes

Category
expert
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

GLM-5.1

1,470.12 rating

Confidence interval: 1,456.51 – 1,483.73

2433 votes

Category
french
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

GLM-5.1

1,465.69 rating

Confidence interval: 1,446.92 – 1,484.45

1090 votes

Category
german
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

GLM-5.1

1,473.04 rating

Confidence interval: 1,468.97 – 1,477.12

43616 votes

Category
hard_prompts
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

GLM-5.1

1,482.51 rating

Confidence interval: 1,477.22 – 1,487.80

18875 votes

Category
hard_prompts_english
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

GLM-5.1

1,453.49 rating

Confidence interval: 1,447.42 – 1,459.56

12773 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

GLM-5.1

1,442.22 rating

Confidence interval: 1,436.46 – 1,447.98

15875 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

GLM-5.1

1,465.74 rating

Confidence interval: 1,457.03 – 1,474.45

5225 votes

Category
industry_legal_and_government
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

GLM-5.1

1,478.55 rating

Confidence interval: 1,472.19 – 1,484.91

10954 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

GLM-5.1

1,471.40 rating

Confidence interval: 1,461.03 – 1,481.77

3647 votes

Category
industry_mathematical
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

GLM-5.1

1,473.73 rating

Confidence interval: 1,464.50 – 1,482.96

4717 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

GLM-5.1

1,483.66 rating

Confidence interval: 1,478.97 – 1,488.34

26126 votes

Category
industry_software_and_it_services
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

GLM-5.1

1,454.52 rating

Confidence interval: 1,449.03 – 1,460.01

17659 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

GLM-5.1

1,452.50 rating

Confidence interval: 1,447.58 – 1,457.42

23325 votes

Category
instruction_following
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

GLM-5.1

1,440.34 rating

Confidence interval: 1,419.24 – 1,461.44

959 votes

Category
japanese
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

GLM-5.1

1,423.16 rating

Confidence interval: 1,405.03 – 1,441.30

1247 votes

Category
korean
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

GLM-5.1

1,467.18 rating

Confidence interval: 1,462.46 – 1,471.90

29944 votes

Category
longer_query
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

GLM-5.1

1,472.86 rating

Confidence interval: 1,462.16 – 1,483.57

3303 votes

Category
math
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

GLM-5.1

1,472.40 rating

Confidence interval: 1,465.87 – 1,478.92

10549 votes

Category
multi_turn
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

GLM-5.1

1,448.65 rating

Confidence interval: 1,444.48 – 1,452.81

38380 votes

Category
non_english
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

GLM-5.1

1,462.82 rating

Confidence interval: 1,459.31 – 1,466.32

66771 votes

Category
overall
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

GLM-5.1

1,451.83 rating

Confidence interval: 1,434.87 – 1,468.79

1207 votes

Category
polish
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

GLM-5.1

1,455.21 rating

Confidence interval: 1,448.02 – 1,462.41

8091 votes

Category
russian
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

GLM-5.1

1,468.57 rating

Confidence interval: 1,455.12 – 1,482.01

2220 votes

Category
spanish
Evaluated configuration
GLM-5.1
Evaluation type
human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating

GLM-5.1

1,514.33 rating

Confidence interval: 1,504.79 – 1,523.87

4393 votes

Category
chinese
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

GLM-5.1

1,485.35 rating

Confidence interval: 1,479.88 – 1,490.82

17133 votes

Category
coding
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

GLM-5.1

1,453.51 rating

Confidence interval: 1,447.14 – 1,459.88

12259 votes

Category
creative_writing
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

GLM-5.1

1,475.08 rating

Confidence interval: 1,470.46 – 1,479.69

26947 votes

Category
english
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

GLM-5.1

1,464.93 rating

Confidence interval: 1,460.30 – 1,469.56

46946 votes

Category
exclude_ties
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

GLM-5.1

1,476.94 rating

Confidence interval: 1,469.11 – 1,484.77

6716 votes

Category
expert
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

GLM-5.1

1,471.82 rating

Confidence interval: 1,457.72 – 1,485.92

2245 votes

Category
french
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

GLM-5.1

1,466.55 rating

Confidence interval: 1,447.17 – 1,485.94

1009 votes

Category
german
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

GLM-5.1

1,472.43 rating

Confidence interval: 1,468.27 – 1,476.58

41196 votes

Category
hard_prompts
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

GLM-5.1

1,481.77 rating

Confidence interval: 1,476.38 – 1,487.16

17950 votes

Category
hard_prompts_english
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

GLM-5.1

1,451.78 rating

Confidence interval: 1,445.60 – 1,457.95

12138 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

GLM-5.1

1,442.60 rating

Confidence interval: 1,436.73 – 1,448.48

14929 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

GLM-5.1

1,465.33 rating

Confidence interval: 1,456.45 – 1,474.22

4966 votes

Category
industry_legal_and_government
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

GLM-5.1

1,477.71 rating

Confidence interval: 1,471.23 – 1,484.19

10348 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

GLM-5.1

1,469.38 rating

Confidence interval: 1,458.70 – 1,480.06

3412 votes

Category
industry_mathematical
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

GLM-5.1

1,473.04 rating

Confidence interval: 1,463.61 – 1,482.47

4478 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

GLM-5.1

1,483.25 rating

Confidence interval: 1,478.48 – 1,488.02

24729 votes

Category
industry_software_and_it_services
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

GLM-5.1

1,454.16 rating

Confidence interval: 1,448.56 – 1,459.75

16617 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

GLM-5.1

1,452.44 rating

Confidence interval: 1,447.44 – 1,457.44

22045 votes

Category
instruction_following
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

GLM-5.1

1,435.55 rating

Confidence interval: 1,413.33 – 1,457.77

851 votes

Category
japanese
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

GLM-5.1

1,422.33 rating

Confidence interval: 1,403.89 – 1,440.77

1192 votes

Category
korean
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

GLM-5.1

1,466.90 rating

Confidence interval: 1,462.10 – 1,471.70

28385 votes

Category
longer_query
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

GLM-5.1

1,473.00 rating

Confidence interval: 1,461.97 – 1,484.03

3088 votes

Category
math
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

GLM-5.1

1,472.41 rating

Confidence interval: 1,465.73 – 1,479.08

9941 votes

Category
multi_turn
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

GLM-5.1

1,448.41 rating

Confidence interval: 1,444.16 – 1,452.66

35940 votes

Category
non_english
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

GLM-5.1

1,462.40 rating

Confidence interval: 1,458.83 – 1,465.97

62889 votes

Category
overall
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

GLM-5.1

1,447.48 rating

Confidence interval: 1,430.26 – 1,464.69

1156 votes

Category
polish
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

GLM-5.1

1,453.86 rating

Confidence interval: 1,446.47 – 1,461.24

7504 votes

Category
russian
Evaluated configuration
glm-5.1
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

GLM-5.1

1,468.54 rating

Confidence interval: 1,454.67 – 1,482.42

2067 votes

Category
spanish
Evaluated configuration
glm-5.1
Evaluation type
human preference

Run it locally

Inference · memory estimate

GLM-5.1

1404.2 GiB checkpoint tensor payload · 753.86 B stored tensor elements · BF16/F32

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.

Capabilities

Context window
202,752 tokens
Maximum input
200,000 tokens*
Maximum output
128,000 tokens*
Reasoning
Yes*
Tool calling
Yes*
Open weights
Yes
API mode
Chat*
Prompt caching
Yes*
Tool choice
Yes*
Model license
mit
Checkpoint tensor payload
1,404.18 GiB

Input modalities

  • text

Output modalities

  • text

Processing & platform

Architecture
glm_moe_dsa

System configuration

Detailed specifications

Specifications · 12 fields
cache creation input token cost
0*
Scope and conditions
native_context
provider
zai*
mode
chat*
full_native_model_id
zai/glm-5.1*
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.*
cache read input token cost
2.6e-7*
Scope and conditions
native_context
provider
zai*
mode
chat*
full_native_model_id
zai/glm-5.1*
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.*
input cost per token
1.4e-6*
Scope and conditions
native_context
provider
zai*
mode
chat*
full_native_model_id
zai/glm-5.1*
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-5.1*
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
200,000*
Scope and conditions
native_context
provider
zai*
mode
chat*
full_native_model_id
zai/glm-5.1*
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
128,000*
Scope and conditions
native_context
provider
zai*
mode
chat*
full_native_model_id
zai/glm-5.1*
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-5.1*
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
4.4e-6*
Scope and conditions
native_context
provider
zai*
mode
chat*
full_native_model_id
zai/glm-5.1*
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-5.1*
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 prompt caching
Yes*
Scope and conditions
native_context
provider
zai*
mode
chat*
full_native_model_id
zai/glm-5.1*
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 reasoning
Yes*
Scope and conditions
native_context
provider
zai*
mode
chat*
full_native_model_id
zai/glm-5.1*
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-5.1*
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.*

* Provisional data