Model lifecycle
DeepSeek V4 Pro
Released 2026-04-24 · Back to DeepSeek
Exact-version signals observed 2026-06-12–2026-07-31 · n=412
Lifecycle updated daily · latest complete UTC day 2026-07-31
- Fixed 14-day baseline
- 66.9 n=65
- Latest 28 days
- 68.3 n=243
- Change vs baseline
- +1.4
- 90-day slope
- −0.7 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.
4 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.
No category has enough samples in both windows for a contribution conclusion.
| Category | Baseline → current | Category change | Weight share | Experience contribution | Discussion-mix contribution |
|---|---|---|---|---|---|
| Speed & latency | 69.4 → 69.4n=12 → 44 | +0.1 | 19.9% → 18.6% | +0.01 | −0.06 |
| General text | 62.4 → 64.2n=28 → 107 | +1.8 | 40.2% → 41.7% | +0.74 | −0.02 |
| Coding | 63.5 → 71.1n=14 → 51 | +7.6 | 21.9% → 21.7% | +1.65 | −0.00 |
| Image / vision | not enough datan=1 → 2 | – | – | – | – |
| Local deploy | not enough datan=1 → 0 | – | – | – | – |
| Reasoning | not enough datan=9 → 39 | – | – | – | – |
| Roleplay / creative | not enough datan=0 → 0 | – | – | – | – |
| Safety & refusals | not enough datan=0 → 1 | – | – | – | – |
Unresolved contribution from categories below the sample threshold: −0.91 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.