#89 GLM 5
none- Cost
- $0.007
- Time
- 32.1s
- Tokens
- 2,023 tok
Z.ai
Release: 2026-02-12
Tested on: 2026-05-21 23:41
z-ai/glm-5::none
744B total (40B active)MoEOpen source
(medium)
(none)
Summary
GLM 5 scores 6.3 on AI BENCHY and ranks #89. It has 10.0 reliability, a 46.7% pass rate, $0.023 total cost, and 3.97s average response time.
What makes GLM 5 unique: Its total benchmark cost is unusually low for its score range. It is notably fast compared with similar models.
Model facts
Researched on 2026-08-12
- Parameters
- 744B total (40B active)
- Architecture
- MoE
- Availability
- Open source
- License
- MIT
6.3
Consistency
9.7
10.0
$0.023
Total Output Tokens
1,988
Total Input Tokens
0
Input Price
$0.600 / 1M
Output Price
$1.920 / 1M
Flaky tests
1
Flaky tests had mixed outcomes across runs (at least one pass and one fail).
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
Run history
| Tested on | Score | Reliability | Tests Correct | Total Cost | Compare |
|---|---|---|---|---|---|
| 2026-06-04 13:05 Re-test | 5.7 | 10.0 | $0.041 ↑ | Compare | |
| 2026-06-04 13:05 Re-test | 5.7 | 10.0 | $0.042 ↑ | Compare | |
| 2026-06-04 13:05 New test added | 6.1 | 10.0 | $0.027 ↑ | Compare | |
| 2026-05-21 23:41 Suite changed | 6.3 | 10.0 | $0.023 | Current run | |
| 2026-04-22 12:55 First recorded run | 6.6 | N/A | $0.020 | Compare |
This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.
Run comparison
| Run | Benchmark coverage | Score | Consistency | Reliability | Tests Correct | Flaky tests | Total Output Tokens | Total Input Tokens | Total Cost | Response Time (avg) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-05-21 23:41 · Suite changed | 60/60 attempts | 6.3 | 9.7 | 10.0 | 9/20 | 1 | 1,988 | 0 | $0.023 | 3.97s |
| 2026-06-04 13:05 · Re-test | 63/63 attempts | 5.7 | 9.3 | 10.0 | 9/21 | 1 | 1,989 | 37,135 | $0.042 | 4.03s |
| Difference | — | +0.6 | +0.4 | 0.0 | 0 | 0 | -1 | -37135 | -$0.020 | -53ms |
Benchmark coverage differs: 60/60 attempts (Target: 3 repeats per test) versus 63/63 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.
These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.
Charts
Choose the first model, then click a second model to open a side-by-side page.
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Quick Compare
GLM 5nonevsNemotron 3 SupermediumFree AvailableGLM 5nonevsInkling SmalllowGLM 5nonevsNorth Mini CodemediumFree AvailableGLM 5nonevsLing-3.0-flashnoneGLM 5nonevsSolar Pro 4noneGLM 5nonevsGLM 5.1noneGLM 5nonevsGPT-5.4noneGLM 5nonevsKimi K2.6noneGLM 5nonevsSeed 2.1 TurbononeGLM 5nonevsQwen3.5-122B-A10Bnone
Category Breakdown
| Category | Score | Consistency | Tests Correct |
|---|---|---|---|
| Anti-AI Tricks | 4.8 | 10.0 | |
| Coding | 4.6 | 6.8 | |
| Combined | 3.0 | 10.0 | |
| Data parsing and extraction | 10.0 | 10.0 | |
| Domain specific | 3.0 | 10.0 | |
| General Intelligence | 10.0 | 10.0 | |
| Instructions following | 10.0 | 10.0 | |
| Puzzle Solving | 7.7 | 10.0 | |
| Tool Calling | 10.0 | 10.0 | |
| Trivia | 3.0 | 10.0 |