Noometry

Z.ai (Zhipu), open weights

GLM-5.2

GLM-5.2 by Z.ai (Zhipu) ranks 44th of 354 ranked models on the Noometry Index as of October 2026, with a score of 51.1. Its strongest category is writing & preference, where it ranks 21st. API pricing starts at $1.40 per million input tokens and $4.40 per million output tokens, with a 1M-token context window.

Last verified

Noometry rank
#44 of 354
Index score
51.1
Evidence
Confirmed 51 results
Provider
Z.ai (Zhipu)
Released
June 13, 2026
Weights
Open weights
Reasoning
Yes
Context window
1M
Max output
131K
Input price
$1.40 / M
Output price
$4.40 / M
Blended price
$2.15 / M
Output speed
23 tokens/s Kagi
Value
#146 of 219
Knowledge cutoff
Unknown
Input
text
Hugging Face
zai-org/GLM-5.2

Category scores

Each category score combines every public result we have in that category.

  1. Coding 51.3
  2. Agentic & Tool Use 32.4
  3. Reasoning 42.3
  4. Math 55.7
  5. Knowledge 57.1
  6. Multilingual 55.8
  7. Instruction Following 76.9
  8. Long Context 45.3
  9. Writing & Preference 70.4
CategoryScoreRankResults
Coding51.3#417
Agentic & Tool Use32.4#634
Reasoning42.3#5213
Math55.7#436
Knowledge57.1#403
Multilingual55.8#261
Instruction Following76.9#341
Long Context45.3#431
Writing & Preference70.4#215

Strengths and weaknesses

Categories where GLM-5.2 places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Writing & Preference70.4+16.7#21 of 312, top 7%
Multilingual55.8+8.4#26 of 297, top 9%
Instruction Following76.9+5.7#34 of 305, top 12%

Weakest categories

CategoryScorevs medianRank
Agentic & Tool Use32.4+2.1#63 of 154, top 41%
Reasoning42.3+18.7#52 of 350, top 15%
Long Context45.3+4.4#43 of 296, top 15%

Closest competitors

The models ranked just above and below GLM-5.2. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Gemini 3 Flash Preview#4052.3$1.13—Compare
GLM-5.3-Flash#4151.8$0.24—Compare
Qwen3.7 Max#4251.5$3.75—Compare
Qwen3.6 Max Preview#4351.5$2.92—Compare
GPT-5#4550.9$3.442Compare
Muse Spark#4650.6——Compare
Claude Opus 4.5#4750.5$1013Compare
Muse Spark 1.2#4850.3$2—Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
SWE-bench Verified78.7%#5 of 32, top 16%maxEpoch AI2026-06-25
DeepSWE36.3%highEpoch AI
DeepSWE43.8%#25 of 29, top 87%maxEpoch AI
FrontierCode24.5%#30 of 37, top 82%noneEpoch AI
LMArena WebDev1603#22 of 113, top 20%LMArena2026-10-08
SciCode50.5%#41 of 121, top 34%maxEpoch AI
SciCode36.1%noneEpoch AI
WeirdML67.3%highEpoch AI
WeirdML70.1%#18 of 119, top 16%maxEpoch AI
LMArena Coding1485#38 of 294, top 13%LMArena2026-10-08
ALE-Bench1,047#40 of 105, top 39%highEpoch AI
ALE-Bench1,010maxEpoch AI

Agentic & Tool Use

BenchmarkScorePositionSettingSourceDate
APEX-Agents45.2%#32 of 49, top 66%Epoch AI
τ²-bench Banking37.1%#11 of 26, top 43%xhighτ²-bench2026-08-04
PostTrainBench31.7%#6 of 11, top 55%maxEpoch AI
GBAEval0%#21 of 23, top 92%Epoch AI
Vending-Bench 28,314#10 of 60, top 17%Epoch AI

Reasoning

BenchmarkScorePositionSettingSourceDate
ARC-AGI-222.8%#41 of 83, top 50%Epoch AI
SimpleBench58.8%#28 of 77, top 37%Epoch AI
Kagi LLM Benchmark62.6%#37 of 99, top 38%Kagi LLM Benchmark
Kagi LLM Benchmark60%Kagi LLM Benchmark
NYT Connections (extended)74.3%#45 of 91, top 50%high reasoningLech Mazur benchmarks
ARC-AGI-177%#40 of 83, top 49%Epoch AI
CritPt20.9%#21 of 134, top 16%maxEpoch AI
CritPt3.1%noneEpoch AI
Chess Puzzles14%lowEpoch AI2026-08-10
Chess Puzzles21%#52 of 129, top 41%maxEpoch AI2026-06-17
Chess Puzzles6%noneEpoch AI2026-08-10
EBR-Bench9.5%#19 of 24, top 80%maxEpoch AI2026-06-29
LMArena Hard Prompts1480#34 of 297, top 12%LMArena2026-10-08
Mystery Game Puzzles19%#44 of 74, top 60%lowEpoch AI2026-08-27
Mystery Game Puzzles15%mediumEpoch AI2026-08-27
Mystery Game Puzzles18%minimalEpoch AI2026-08-27
Mystery Game Puzzles18%noneEpoch AI2026-08-27
DTBench93.6%#25 of 151, top 17%maxEpoch AI
LMCA45.8%#31 of 125, top 25%maxEpoch AI
Surface Evolver Bench55.6%#12 of 25, top 48%highEpoch AI
Epoch Capabilities Index151.78#48 of 213, top 23%Epoch AI2026-06-16

Math

BenchmarkScorePositionSettingSourceDate
FrontierMath (Tiers 1-3)54.7%lowEpoch AI2026-08-29
FrontierMath (Tiers 1-3)59.2%#36 of 81, top 45%maxEpoch AI2026-06-19
FrontierMath (Tiers 1-3)42.5%noneEpoch AI2026-08-29
FrontierMath Tier 429.3%#30 of 63, top 48%maxEpoch AI2026-06-19
MathArena Final-Answer Competitions67.6%#16 of 29, top 56%MathArena
OTIS Mock AIME 2024-202575.6%lowEpoch AI2026-08-10
OTIS Mock AIME 2024-202586.4%#67 of 173, top 39%maxEpoch AI2026-06-25
OTIS Mock AIME 2024-202528.9%noneEpoch AI2026-08-10
ProofBench35%#39 of 77, top 51%maxEpoch AI
LMArena Math1482#26 of 285, top 10%LMArena2026-10-08

Knowledge

BenchmarkScorePositionSettingSourceDate
GPQA Diamond87.9%lowEpoch AI2026-08-10
GPQA Diamond91.9%#23 of 186, top 13%maxEpoch AI2026-06-24
GPQA Diamond71.2%noneEpoch AI2026-08-10
SimpleQA Verified34.2%#50 of 77, top 65%maxEpoch AI2026-08-27
LMArena Expert1486#36 of 273, top 14%LMArena2026-10-08

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1459#26 of 297, top 9%LMArena2026-10-08
LMArena Chinese1519#27 of 285, top 10%LMArena2026-10-08
LMArena French1479#27 of 223, top 13%LMArena2026-10-08
LMArena German1468#29 of 231, top 13%LMArena2026-10-08
LMArena Japanese1451#25 of 211, top 12%LMArena2026-10-08
LMArena Korean1445#22 of 213, top 11%LMArena2026-10-08
LMArena Russian1466#30 of 283, top 11%LMArena2026-10-08
LMArena Spanish1477#15 of 226, top 7%LMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LMArena Instruction Following1465#30 of 298, top 11%LMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1479#25 of 291, top 9%LMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1470#27 of 297, top 10%LMArena2026-10-08
LMArena Creative Writing1462#18 of 295, top 7%LMArena2026-10-08
EQ-Bench Creative Writing1757#26 of 115, top 23%EQ-Bench
EQ-Bench 41222#14 of 28, top 50%EQ-Bench
LMArena Multi-Turn1469#31 of 295, top 11%LMArena2026-10-08

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
alibaba$1.40$4.40$0.282026-10-10
deepinfra$0.75$2.40$0.142026-10-10
mistral$1.40$4.40$0.142026-10-10
openrouter$0.06$7$0.0592026-10-10
together$1.40$4.40$0.262026-10-10
vertex$1.40$4.40$0.142026-10-10
zai$1.40$4.40$0.262026-10-10

All Z.ai (Zhipu) API prices → Estimate your cost →

Compare GLM-5.2

Other Z.ai (Zhipu) models

Frequently asked questions

How good is GLM-5.2?

GLM-5.2 by Z.ai (Zhipu) ranks 44th of 354 ranked models on the Noometry Index as of October 2026, with a score of 51.1. Its strongest category is writing & preference, where it ranks 21st. API pricing starts at $1.40 per million input tokens and $4.40 per million output tokens, with a 1M-token context window.

How much does GLM-5.2 cost?

GLM-5.2 costs $1.40 per million input tokens and $4.40 per million output tokens on Z.ai (Zhipu)'s own API, with cached input at $0.26.

What is GLM-5.2's context window?

GLM-5.2 accepts up to 1M tokens of input and can write up to 131K tokens in one response.

Is GLM-5.2 open source?

Yes. GLM-5.2's weights are downloadable from Hugging Face (zai-org/GLM-5.2); check the license for commercial terms.

How fast is GLM-5.2?

GLM-5.2 generated about 23 output tokens per second in the Kagi LLM Benchmark's timed runs. Speed varies by provider, load and reasoning effort.

What are GLM-5.2's strengths and weaknesses?

Relative to other ranked models, GLM-5.2 places best in writing & preference, multilingual, instruction following and lowest in agentic & tool use, reasoning, long context.

What is GLM-5.2 best at?

Its best category is writing & preference, where it ranks 21st on Noometry.