Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage & billing
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Usage & billing
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksNano RPCEarn crypto
  1. Models
  2. Text
  3. Z.AI

GLM 5.2 Thinking

GLM-5.2 with thinking enabled for harder long-horizon coding, autonomous agent workflows, complex engineering optimization, and real-world development tasks.

  • Reasoning
  • Tool Calling
  • Structured Output

Added Jun 15, 2026

Model weights

Pricing

Auto routing · per 1M tokens
Input
$0.42
Output
$1.32
Cache read
$0.078
Compare provider prices

Specifications

Context window
1M
Max output
131.1K
Parameters
744B / 40B
Total / active

Benchmarks

Sourced from LMArena.

Arena Score

1475.0

Overall Rank

#39 / 421

Votes

54,143

Confidence Interval

1470.9 - 1479.1

Category Scores

Coding

#48 / 416

14,364 votes

1515.4

Math

#31 / 404

2,415 votes

1486.9

Longer Query

#37 / 399

24,486 votes

1485.7

Creative Writing

#39 / 419

10,851 votes

1450.6

Instruction Following

#41 / 421

18,594 votes

1470.0

Hard Prompts

#44 / 421

35,069 votes

1495.9

Additional Categories
22

Spanish

#14 / 304

1,663 votes

1483.7

Industry Legal And Government

#29 / 393

4,342 votes

1494.3

Industry Life And Physical And Social Science

#29 / 419

8,953 votes

1497.4

Industry Mathematical

#30 / 402

2,871 votes

1493.3

Industry Medicine And Healthcare

#32 / 391

3,960 votes

1491.9

Chinese

#33 / 395

3,861 votes

1523.3

English

#37 / 421

21,986 votes

1482.6

Exclude Ties

#40 / 421

40,217 votes

1484.4

Japanese

#40 / 288

892 votes

1456.1

French

#42 / 301

2,031 votes

1489.0

German

#42 / 320

991 votes

1473.7

Industry Writing And Literature And Language

#42 / 420

14,509 votes

1458.4

Non English

#42 / 421

32,150 votes

1461.7

Industry Software And It Services

#43 / 421

20,850 votes

1507.6

Expert

#44 / 371

5,520 votes

1501.9

Hard Prompts English

#44 / 419

14,479 votes

1498.1

Korean

#44 / 289

978 votes

1430.4

Industry Entertainment And Sports And Media

#45 / 419

13,425 votes

1445.7

Russian

#45 / 385

6,534 votes

1472.3

Multi Turn

#49 / 419

8,616 votes

1476.5

Industry Business And Management And Financial Operations

#55 / 414

10,290 votes

1464.9

Polish

#58 / 243

904 votes

1463.7

Published 2026-11-11 · Matched as GLM-5.2 (Max) · CC BY 4.0

LMArena Dataset

Leaderboard results

BenchmarkScoreRank on NanoGPT
Epoch AI Benchmarking Hub

Epoch AI data, CC BY 4.0 · Updated Nov 18, 2026

Epoch Capabilities Index

Tested as GLM 5.2

153.0

#50 of 113

FrontierMath v2 (Tiers 1–3)

59.2%

#42 of 82

SWE-bench Verified

78.7%

#4 of 27

GPQA Diamond

91.9%

#27 of 117

SimpleQA Verified

34.2%

#58 of 82

FrontierMath v2 (Tier 4)

29.3%

#35 of 66

OTIS Mock AIME 2024–2025

86.4%

#72 of 119

MirrorCode

36.7%

#28 of 40

Chess Puzzles

22.0%

#56 of 81

Mystery Game Puzzles

Tested as GLM 5.2

19.0%

#52 of 78

Sierra τ-bench

tau2-bench data, MIT · Updated Aug 4, 2026

τ³-Banking

Tested as GLM 5.2

37.1%

#10 of 18

EQ-Bench

EQ-Bench data, MIT · Updated Nov 18, 2026

EQ-Bench 4

Tested as GLM 5.2

1208

#18 of 31

Creative Writing v3

Tested as GLM 5.2

1759

#28 of 91

Longform Writing

Tested as GLM 5.2

77.9

#24 of 93

EQ-Bench 3

Tested as GLM 5.2

1563

#6 of 45

UGI Leaderboard

UGI Leaderboard data, Apache-2.0 · Updated Oct 16, 2026

UGI Score

Tested as GLM 5.2

58.9

#8 of 125

UGI Willingness

Tested as GLM 5.2

2.8

#66 of 125

UGI Writing

Tested as GLM 5.2

67.0

#18 of 125

Filtered and ranked to models on NanoGPT.

Providers

Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)

Auto
Auto (ZDR)
$0.50$3.15$0.078N/AN/A
Auto
Auto (Anonymized)
$0.42$1.32$0.078N/AN/A
Relace
Relace
FP4
UNK
$0.00693$8.38$0.00693N/AN/A
Wafer US
Wafer US
FP4
US
$0.00735$7.14$0.0063N/AN/A
Cloudflare
Cloudflare
US
$0.20$8.40$0.20N/AN/A
Wafer
Wafer
FP4
US
$0.20$8.40$0.189N/AN/A
DeepInfra
DeepInfra
FP4
US
$0.591$1.89$0.11N/AN/A
Ambient
Ambient
FP8
$0.63$2.10$0.157N/AN/A
SCX (EU only)
SCX (EU only)
BF16
EU
$0.635$2.23$0.162N/AN/A
SCX (Global)
SCX (Global)
BF16
GLB
$0.635$2.23$0.162N/AN/A
StreamLake
StreamLake
56% off
FP8
CN
$1.47$0.644$4.62$2.02$0.273$0.12N/AN/A
CoreWeave
CoreWeave
FP4
US
$0.651$2.54$0.147N/AN/A
Morph
Morph
FP8
US
$0.675$2.08$0.228N/AN/A
SiliconFlow
SiliconFlow
FP8
SG
$0.684$2.15$0.136N/AN/A
EngyAI
EngyAI
GLB
$0.714$1.57$0.189N/AN/A
Runware
Runware
10% off
UNK
$0.84$0.756$2.68$2.41$0.168$0.151N/AN/A
Alibaba
Alibaba
FP8
SG
$1.01$3.19$0.203N/AN/A
Aster
Aster
$1.05$4.20$0.21N/AN/A
MegaNova
MegaNova
FP8
US
$1.05$3.15$0.189N/AN/A
Parasail
Parasail
FP4
US
$1.16$3.47$0.21N/AN/A
TensorX
TensorX
FP8
EU
$1.16$3.47$0.289N/AN/A
Uomi
Uomi
BF16
GLB
$1.37$4.30N/AN/AN/A
AtlasCloud
AtlasCloud
FP8
US
$1.47$4.62$0.273N/AN/A
Baidu
Baidu
FP8
CN
$1.47$4.62$0.273N/AN/A
BaseTen
BaseTen
FP8
US
$1.47$4.62$0.147N/AN/A
Friendli
Friendli
US
$1.47$4.62$0.273N/AN/A
GMICloud
GMICloud
FP8
US
$1.47$4.62$0.273N/AN/A
Kitani
Kitani
GLB
$1.47$4.62$0.273N/AN/A
Nebius
Nebius
FP4
GLB
$1.47$4.62$0.147N/AN/A
Novita
Novita
FP8
US
$1.47$4.62$0.273N/AN/A
Ollama
Ollama
UNK
$1.47$4.62$0.273N/AN/A
Together
Together
US
$1.47$4.62$0.273N/AN/A
Venice
Venice
FP8
US
$1.47$4.62$0.273N/AN/A
Z.AI
Z.AI
FP8
SG
$1.47$4.62$0.273N/AN/A
Privacy:ZDR, prompts not retainedNo training on promptsRetention and training unclearMay train on prompts
Batch API
ProviderInputOutputCached input
Auto
Autoz-ai/glm-5.2:thinking
$0.42/M$1.32/M$0.078/M

Batch rows for this model use normal routing and cost the same as standard API requests. Set provider in a row to choose a provider.

Related text models

Compare GLM 5.2 Thinking with similar models from the same provider or model family.

GLM 5.3 FlashX Legal

z-ai/glm-5.3-flashx-legal

GLM 5.3 FlashX Legal is Z.AI's model trained for legal document analysis, contract review, legal reasoning and initial drafts. It accepts text and images and returns text, with always-on reasoning, tool calling, JSON output, a 1M-token context window and up to 131,072 output tokens. External legal databases and document-editing tools require your own integrations.

Apex Flash 1

z-ai/glm-5.3-flash-apex

Apex Flash 1 is Cantina Security and Yeta's cybersecurity-focused post-train of GLM 5.3 Flash for code investigation and tool use.

Apex Flash 1 Abliterated

z-ai/glm-5.3-flash-apex-abliterated

Apex Flash 1 Abliterated is an experimental derivative of Cantina Security and Yeta's Apex Flash 1 with modified refusal behavior.

GLM 5.3 Uncensored

z-ai/glm-5.3-uncensored

An experimental uncensored version of the full GLM 5.3 reasoning model for chat, creative writing, coding, and tool use. It may show unusual behavior or inconsistent response quality.

GLM 5.3 Flash Cybersecurity

z-ai/glm-5.3-flash-cybersecurity

GLM 5.3 Flash Cybersecurity is a cybersecurity-focused variant based on the uncensored model, with provider moderation for illegal activities. It supports always-on reasoning, image understanding, tool calling, and a 1,048,576-token context window.

GLM 5.3 Flash

z-ai/glm-5.3-flash

ox-alpha out of stealth! GLM-5.3 Flash is Z.ai's first natively multimodal GLM-5 model, with 320B total parameters and just 18B active parameters for efficient coding, agentic work, and precise 1M-token context. Its hybrid sparse-and-linear attention architecture helps it outperform GLM-5.2 at one-tenth the price while approaching Claude Opus 4.8 on coding and agentic benchmarks.