I-129F Final-Stage Median Falls 25 Days, But Recent Filers Face a Different Reality

The week of August 17 produced a 155-day final-stage median for I-129F approvals, down from 180 days the prior week, a 25-day single-week drop. Compared with the same week a year ago (261 days), the improvement is 40.5%, though the prior-year figure is based on only 18 cases and should be read with caution. The 3-year same-week average of 261 days is a more stable baseline, and this week's reading sits 106 days below it. The 12-week trailing average is 175 days, so this week is 20 days below that pace as well.

The key tension in this week's data is the difference between what the final-stage median measures and what recent filers are experiencing. The 155-day figure reflects cases approved this week, meaning cases that entered the pipeline many months ago. The filing-month cohort (a group of cases) burndown (cases filed in the same calendar month, tracked through approval) tells a different story: all six filing cohorts from March through August 2026 show 0.0% approved as of the August 26 snapshot. The cases clearing now are not the cases recently filed.

Over the past 12 weeks, I-129F final-stage medians have ranged from 146 days (week of June 29) to 201 days (week of August 3), a 55-day swing. The 8-week regression slope is +0.5 days per week, and the R-squared is 0.00, meaning the linear fit explains essentially none of the week-to-week variation. The pattern is classified as noisy_stable. This week's 155-day reading is consistent with that oscillation: it is near the lower end of the range, but so was June 29's 146-day reading, which was followed by a return to 197 days just two weeks later.

The series has swung 55 days in 12 weeks; the direction is not the signal, the volatility is.

The median (the 50th percentile, meaning half of cases finish faster and half slower) is the primary metric here. It is less sensitive to outlier cases than an average would be, but in a noisy series like this one, even the median reflects who happened to be approved in a given week as much as any underlying shift. The prior-year same-week comparison (261 days, n = 18) is statistically thin; the 3-year same-week average of 261 days provides a more grounded reference point, and the 106-day gap versus this week's reading is the more credible year-over-year signal.

This week's data includes an anomaly worth understanding: cases in which a petitioner responded to an RFE (Request for Evidence, USCIS's standard follow-up letter asking for missing documents) cleared in a median of just 14 days, compared with 256 days for cases that went through plain processing without an RFE. That is an 18x difference and the trigger for the rfe_fast anomaly flag in the weekly detector. For reference, the stage-duration data (raw percentiles across the 30-day window) shows RFE-response cases at a median of 21 days across 5,511 cases, consistent with a cluster of responses moving through quickly.

Receiving an RFE does not by itself signal a denial; most cases that receive one are still approved. The unusually fast clearance this week most likely reflects a small group of RFE responses filed around the same time moving through the adjudication (USCIS's case decision step) queue together, what the data community calls a batch effect. The rfe_fast flag is rare precisely because this pattern is atypical. Whether it repeats in coming weeks is one of the signals worth monitoring; a second consecutive fast week would be more meaningful than a single observation.