Gate: computed weekly across the cohort; no convergence event this week
Gate: computed weekly across the cohort; no convergence event this week
position in cohort distributionone of 7 in cohort
percentile
p83
momentum averages present-axis z-scores; two- and three-axis scores are not strictly comparable
velocity z
velocity z
+1.30 σ
slope
+0.013/wk
log-slope +0.013/wk
citation z
∅no citation axis
Citation axis: insufficient source coverage (source returned too few observations).
direct-dependents z
direct-dependents z
+0.00 σ
slope
+0.007/point
Direct-dependents axis: measured.
cohort
7
systems measured
weekly commits
0.4
Development velocity axis: measured.
total citations
n/a
Citation axis: insufficient source coverage (source returned too few observations).
momentum
58.1
percentile p83
commit heatmap, last 26 weeks0.4/wk
weeks
26
peak/wk
5
active
3
total
7
citations by yearCitation axis: insufficient source coverage (source returned too few observations).
∅Citation axis: insufficient source coverage (source returned too few observations).
UI-TARS Desktop holds an Evidaxis momentum score of 58.1 of 100 on methodology m3, snapshot 2026-09-26, measured within the GUI & Computer-Use Agents cohort (n = 7). Axis coverage: 2 of 3 axes measurable. Gate: computed weekly across the cohort; no convergence event this week
Derived signals
second-order reads, computed from the published series
Three signals ship when the commit series is statistically sufficient; the rest are held in a reserved state
until the data supports them. A reserved signal shows its reason, never a fabricated number.
How these are computed.
swing
-0.81×
regime
recent slope sign differs from prior window
windows
13w vs 39w
snapshot 1 of 1 ·· m3
commit Sharpereserved: <26 non-zero commit weeks
changepoint
none detected
snapshot 1 of 1 ·· m3
code-leads-citation lagreserved
reserved: citation series too short / axis absent for cross-correlation
status
reserved
value
insufficient data
snapshot 1 of 1 ·· m3
axis alpha vs cohortreserved
reserved: cohort N too small for regression (power ~0)
status
reserved
value
insufficient data
snapshot 1 of 1 ·· m3
gate ETAreserved
reserved: needs snapshot history (1 snapshot exists)
status
reserved
value
insufficient history
snapshot 1 of 1 ·· m3
methodology gate
snapshot 1 of 1 ·· m3
commit weeks
52
citation years
2
shipped
2
reserved
4
shipped
recency-swing
commit-changepoint
reserved
commit-sharpe
code-leads-citation-lag
axis-alpha
gate-eta
snapshot 1 of 1 ·· m3
Reconstructed history
min
1 commits/wk
max
14 commits/wk
latest
1 commits/wk
points
17
Weekly commits over the past year, reconstructed from git history (GitHub commit_activity).
This is reconstructed, not a point-in-time capture. Evidaxis captures the Type-2 signals
(watchers, issues, dependents) live and never backfills them; those are the moat.
Measurement states
UI-TARS Desktop is under weekly measurement. History: 13 weekly observations; minimum 4. Development velocity axis: measured. Citation axis: insufficient source coverage (source returned too few observations). Direct-dependents axis: measured. Gate: computed weekly across the cohort; no convergence event this week
This axis measures the frozen-panel trajectory of the package set selected and linkage-verified as of 2026-07-21. Systems outside that panel are not measured on this axis. Residual survivorship beyond the frozen universe is disclosed, not denied.
The earliest weekly points of each series were read from deps.dev's BigQuery history on 2026-07-21; later points are weekly captures. Each record reports how many of its points are reconstructed history.
Source: deps.dev (Google), CC-BY 4.0. Modified: aggregated to system-level counts and momentum scores.
Score derived from public data on Evidaxis methodology m3, never edited by hand. Cite as:
“Evidaxis, UI-TARS Desktop, entity e_A4CCB1NFPZB, snapshot 2026-w39.”