Model lifecycle
GLM 5.2
Released 2026-06-16 · Back to GLM
Exact-version signals observed 2026-06-16–2026-07-24 · n=1,457
Lifecycle updated daily · latest complete UTC day 2026-07-24
- Fixed 14-day baseline
- 66.5 n=756
- Latest 28 days
- 67.4 n=851
- Change vs baseline
- +0.9
- 90-day slope
- – pts / 30 days
14-day rolling experience
Each daily point summarizes the previous 14 complete UTC days. The dashed line is the fixed release baseline; the verdict still uses independent non-overlapping 14-day segments.
2 of 4 independent 14-day segments ready for trend judgment.
What explains the change
Experience change and discussion-mix change are separated. These are observational contributions, not proof of cause.
Largest negative experience contributions: Reasoning, Coding.
| Category | Baseline → current | Category change | Weight share | Experience contribution | Discussion-mix contribution |
|---|---|---|---|---|---|
| Reasoning | 85.7 → 79.8n=178 → 177 | −6.0 | 25.4% → 22.0% | −1.42 | −0.56 |
| Coding | 78.5 → 72.7n=108 → 96 | −5.9 | 15.0% → 12.0% | −0.79 | −0.27 |
| Speed & latency | 47.8 → 50.3n=198 → 211 | +2.5 | 25.4% → 24.8% | +0.62 | +0.12 |
| Local deploy | 29.7 → 51.0n=44 → 47 | +21.3 | 5.5% → 5.2% | +1.14 | +0.08 |
| General text | 67.0 → 71.2n=214 → 306 | +4.2 | 26.8% → 34.3% | +1.28 | +0.20 |
| Image / vision | not enough datan=7 → 1 | – | – | – | – |
| Roleplay / creative | not enough datan=2 → 4 | – | – | – | – |
| Safety & refusals | not enough datan=6 → 10 | – | – | – | – |
Unresolved contribution from categories below the sample threshold: +0.47 pts.
Lifecycle scores use the same experience-signal weights as the main index, without sample shrinkage after n=30. They measure public user perception, not model capability or backend causes.