GLM 5 Original
GLM-5 is a capable open-weight model from Zhipu with advanced reasoning and instruction following. This Original variant uses the model maker's direct endpoint.
- Reasoning
- Tool Calling
- Structured Output
Added Feb 11, 2026
Pricing
Auto routing · per 1M tokens- Input
- $0.60
- Output
- $1.90
- Cache read
- $0.10
Specifications
- Context window
- 200K
- Max output
- 128K
- Parameters
- 744B / 40B
- Total / active
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1457.3
Overall Rank
#67 / 421
Votes
28,979
Confidence Interval
1453.1 - 1461.6
Category Scores
Coding
#86 / 416
7,538 votes
1498.3
Math
#90 / 404
1,681 votes
1442.7
Longer Query
#66 / 399
10,987 votes
1468.6
Creative Writing
#48 / 419
4,814 votes
1446.8
Instruction Following
#73 / 421
9,485 votes
1447.0
Hard Prompts
#71 / 421
18,017 votes
1477.9
Additional Categories22
Chinese
#39 / 395
2,000 votes
1517.3
Korean
#55 / 289
544 votes
1423.6
Industry Entertainment And Sports And Media
#58 / 419
5,842 votes
1437.8
English
#60 / 421
12,764 votes
1471.4
Multi Turn
#60 / 419
4,657 votes
1471.6
Industry Medicine And Healthcare
#61 / 391
2,112 votes
1478.2
Spanish
#63 / 304
1,121 votes
1458.9
Industry Legal And Government
#64 / 393
2,258 votes
1468.7
Industry Writing And Literature And Language
#64 / 420
6,950 votes
1447.3
Hard Prompts English
#65 / 419
8,183 votes
1487.0
Industry Life And Physical And Social Science
#66 / 419
4,675 votes
1478.9
Exclude Ties
#68 / 421
20,991 votes
1461.1
French
#70 / 301
906 votes
1476.7
German
#70 / 320
504 votes
1454.6
Expert
#76 / 371
2,571 votes
1481.7
Industry Business And Management And Financial Operations
#77 / 414
5,864 votes
1451.7
Industry Mathematical
#78 / 402
1,577 votes
1456.3
Japanese
#78 / 288
274 votes
1415.2
Non English
#78 / 421
16,215 votes
1439.3
Russian
#82 / 385
3,417 votes
1448.1
Industry Software And It Services
#83 / 421
11,042 votes
1488.0
Polish
#93 / 243
601 votes
1441.0
Published 2026-11-11 · Matched as GLM-5 · CC BY 4.0
LMArena DatasetLeaderboard results
These leaderboards tested GLM 5, 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 | 146.5 | #76 of 113 |
SWE-bench Verified | 72.1% | #19 of 27 |
GPQA Diamond | 87.8% | #57 of 117 |
OTIS Mock AIME 2024–2025 | 80.0% | #84 of 119 |
Sierra τ-bench tau2-bench data, MIT · Updated Aug 4, 2026 | ||
τ²-bench Overall | 81.0% | #5 of 6 |
Creative Writing v3 | 1601 | #44 of 91 |
Longform Writing | 70.9 | #39 of 93 |
EQ-Bench 3 | 1495 | #21 of 45 |
UGI Leaderboard UGI Leaderboard data, Apache-2.0 · Updated Oct 16, 2026 | ||
UGI Score | 54.6 | #23 of 125 |
UGI Willingness | 1.2 | #119 of 125 |
UGI Writing | 55.0 | #46 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)
| $1.00 | $3.20 | $0.20 | N/A | N/A | |
| $0.60 | $1.90 | $0.10 | N/A | N/A | |
| $0.63 | $2.02 | $0.126 | N/A | N/A | |
| $0.735 | $2.35 | $0.147 | N/A | N/A | |
| $0.921 | $2.87 | N/A | N/A | N/A | |
| $0.998 | $2.68 | $0.21 | N/A | N/A | |
| $1.05 | $3.36 | N/A | N/A | N/A | |
| $1.05 | $3.36 | $0.21 | N/A | N/A | |
| $1.05 | $3.36 | $0.21 | N/A | N/A |
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.