GLM 5.3 Flash TEE
GLM-5.3 Flash is Z.AI's natively multimodal 320B MoE reasoning model with 18B active parameters. This TEE deployment is verified through Redpill attestation with signed completion receipts.
- Reasoning
- Vision
- Tool Calling
- Parallel Tool Calls
- Structured Output
Added Aug 28, 2026
Model weightsPricing
Auto routing · per 1M tokens- Input
- $0.15
- Output
- $0.50
- Cache read
- $0.03
Specifications
- Context window
- 1M
- Max output
- 131.1K
- Parameters
- 320B / 18B
- Total / active
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1475.3
Overall Rank
#36 / 421
Votes
51,019
Confidence Interval
1470.8 - 1479.7
Category Scores
Coding
#28 / 416
12,885 votes
1526.3
Math
#11 / 404
2,537 votes
1504.8
Longer Query
#52 / 399
22,349 votes
1479.7
Creative Writing
#67 / 419
10,973 votes
1434.8
Instruction Following
#35 / 421
17,601 votes
1473.6
Hard Prompts
#35 / 421
32,572 votes
1500.4
Additional Categories22
Spanish
#25 / 304
1,909 votes
1475.1
Chinese
#26 / 395
3,886 votes
1528.6
Industry Software And It Services
#26 / 421
18,959 votes
1515.0
French
#29 / 301
1,914 votes
1498.9
Industry Medicine And Healthcare
#30 / 391
3,402 votes
1493.4
Expert
#31 / 371
4,968 votes
1510.9
Hard Prompts English
#34 / 419
11,880 votes
1501.8
Industry Mathematical
#35 / 402
2,862 votes
1489.9
Japanese
#37 / 288
943 votes
1460.4
Non English
#38 / 421
32,717 votes
1463.9
English
#40 / 421
18,299 votes
1481.5
Industry Life And Physical And Social Science
#41 / 419
8,211 votes
1490.6
Industry Business And Management And Financial Operations
#44 / 414
9,129 votes
1470.8
Exclude Ties
#45 / 421
37,816 votes
1483.6
Korean
#45 / 289
832 votes
1430.3
Russian
#47 / 385
7,564 votes
1471.7
Multi Turn
#51 / 419
7,811 votes
1476.3
Industry Legal And Government
#52 / 393
4,007 votes
1475.3
Industry Writing And Literature And Language
#55 / 420
14,287 votes
1452.3
Industry Entertainment And Sports And Media
#56 / 419
12,959 votes
1438.8
German
#64 / 320
865 votes
1459.4
Polish
#72 / 243
897 votes
1455.4
Published 2026-11-11 · Matched as GLM-5.3-Flash · CC BY 4.0
LMArena DatasetLeaderboard results
These leaderboards tested GLM 5.3 Flash, another version of this model.
| Benchmark | Score | Rank on NanoGPT |
|---|---|---|
Epoch AI Benchmarking Hub Epoch AI data, CC BY 4.0 · Updated Nov 18, 2026 | ||
Epoch Capabilities Index | 153.8 | #51 of 116 |
FrontierMath v2 (Tiers 1–3) | 55.8% | #50 of 84 |
GPQA Diamond | 90.2% | #44 of 120 |
FrontierMath v2 (Tier 4) | 17.1% | #53 of 68 |
OTIS Mock AIME 2024–2025 | 93.9% | #49 of 122 |
Chess Puzzles | 21.0% | #63 of 83 |
Mystery Game Puzzles | 8.0% | #72 of 78 |
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)
| $0.15 | $0.50 | $0.03 | N/A | N/A | |
| $0.15 | $0.50 | $0.03 | N/A | N/A | |
| $0.157 | $0.525 | $0.0315 | N/A | N/A |