GLM 4.5 Air (Thinking)
GLM-4.5-Air with thinking mode enabled for enhanced reasoning capabilities. Shows step-by-step thought process.
- Reasoning
- Tool Calling
- Structured Output
Pricing
Auto routing · per 1M tokens- Input
- $0.13
- Output
- $0.85
- Cache read
- $0.025
Specifications
- Context window
- 128K
- Max output
- 98.3K
- Parameters
- 106B / 12B
- Total / active
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1373.0
Overall Rank
#201 / 421
Votes
30,591
Confidence Interval
1368.7 - 1377.3
Category Scores
Coding
#196 / 416
6,092 votes
1427.0
Math
#186 / 404
1,521 votes
1386.2
Longer Query
#197 / 399
6,755 votes
1379.5
Creative Writing
#210 / 419
3,985 votes
1327.3
Instruction Following
#200 / 421
8,180 votes
1360.8
Hard Prompts
#200 / 421
14,856 votes
1390.7
Additional Categories22
Japanese
#155 / 288
845 votes
1329.8
French
#177 / 301
519 votes
1395.9
German
#178 / 320
743 votes
1355.6
Chinese
#182 / 395
2,230 votes
1405.9
Spanish
#182 / 304
823 votes
1376.0
Industry Mathematical
#186 / 402
1,452 votes
1390.6
Korean
#190 / 289
672 votes
1291.2
Polish
#191 / 243
1,861 votes
1342.3
Industry Life And Physical And Social Science
#193 / 419
4,908 votes
1397.2
Industry Medicine And Healthcare
#194 / 391
1,640 votes
1395.5
Industry Software And It Services
#194 / 421
10,437 votes
1415.2
Industry Business And Management And Financial Operations
#195 / 414
5,491 votes
1372.2
English
#196 / 421
13,183 votes
1390.1
Russian
#198 / 385
1,996 votes
1361.2
Expert
#199 / 371
1,461 votes
1389.8
Hard Prompts English
#199 / 419
6,562 votes
1402.3
Multi Turn
#201 / 419
4,987 votes
1367.3
Non English
#202 / 421
17,366 votes
1352.6
Exclude Ties
#203 / 421
21,487 votes
1343.3
Industry Entertainment And Sports And Media
#210 / 419
5,677 votes
1325.1
Industry Legal And Government
#211 / 393
1,902 votes
1371.5
Industry Writing And Literature And Language
#216 / 420
6,966 votes
1340.6
Published 2026-11-11 · Matched as GLM-4.5-Air · CC BY 4.0
LMArena DatasetLeaderboard results
These leaderboards tested GLM 4.5 Air, another version of this model.
| Benchmark | Score | Rank on NanoGPT |
|---|---|---|
UGI Leaderboard UGI Leaderboard data, Apache-2.0 · Updated Oct 16, 2026 | ||
UGI Score | 46.5 | #51 of 125 |
UGI Willingness | 7.0 | #8 of 125 |
UGI Writing | 47.0 | #62 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)
| $0.20 | $1.10 | $0.03 | N/A | N/A | |
| $0.13 | $0.85 | $0.025 | N/A | N/A | |
| $0.136 | $0.893 | $0.0262 | N/A | N/A | |
| $0.147 | $0.903 | N/A | N/A | N/A | |
| $0.21 | $1.16 | $0.0315 | N/A | N/A |