Evidaxis

SWE-bench

repo incumbent
Coding Agents · n = 20 · SWE-bench/SWE-bench
Calibration · conf low
momentum
momentum 53 of 100 53 /100
momentum
53/100
pctile
p63
convergence gate
Convergence gate: velocity only, citation axis absent × ∅
gate
0 of 2 axes

velocity only, citation axis absent.

gate shut
position in cohort distribution one of 20 in cohort
percentile 63 within cohort p25 p50 p75 p0 p100
percentile
p63
momentum is the combined velocity and citation signal, percentile within cohort
velocity z
velocity z gaugevelocity z+0.24 within cohort, log-slope +0.010/wk.0+0.24
velocity z
+0.24 σ
slope
+0.010/wk
log-slope +0.010/wk
convergence planefull gate = rising
SWE-bench on the convergence plane SWE-bench at velocity z +0.24 and citation z no axis, against 0 dual-axis cohort peers. single-axis: citation axis absent, excluded from the plane. z>=1 median velocity z → citation z ↑
velocity z
+0.24
citation z
no axis
peers
19

single-axis: citation axis absent, excluded from the plane.

citation z no citation axis
cohort
20
systems measured
weekly commits
0.3
not rising
total citations
n/a
no axis
momentum
53.0
percentile p63
commit heatmap, last 26 weeks0.3/wk
weekly commits, 26 weeks26 weeks of commits, peak 4 in a week, 2 active weeks, 5 total.week 1: 0 commitsweek 2: 0 commitsweek 3: 0 commitsweek 4: 0 commitsweek 5: 0 commitsweek 6: 1 commitsweek 7: 0 commitsweek 8: 0 commitsweek 9: 0 commitsweek 10: 0 commitsweek 11: 0 commitsweek 12: 0 commitsweek 13: 0 commitsweek 14: 0 commitsweek 15: 0 commitsweek 16: 0 commitsweek 17: 0 commitsweek 18: 0 commitsweek 19: 0 commitsweek 20: 0 commitsweek 21: 0 commitsweek 22: 0 commitsweek 23: 0 commitsweek 24: 0 commitsweek 25: 0 commitsweek 26: 4 commits
weeks
26
peak/wk
4
active
2
total
5
citations by yearno axis
adoption 7 dependents 6 direct, 1 indirect · pypi/swebench · deps.dev · adoption signal, not part of the momentum score

SWE-bench holds an Evidaxis momentum score of 53.0 of 100 on methodology m2, snapshot 2026-08-08, measured within the Coding Agents cohort (n = 20). It is a mature incumbent included to calibrate the cohort, measured but not gate-eligible.

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.

recency-weighted swing trailing 13-week vs prior 39-week log-slope ratio -1.78, regime reversing. prior slope -0.030, recent slope +0.053 per week. -0.030prior 39w+0.053recent 13w-1.78×
swing
-1.78×
regime
reversing
windows
13w vs 39w
snapshot 1 of 1 · · m2
commit Sharpereserved: <26 non-zero commit weeks
commit changepoint52 weekly commit points. No significant changepoint detected (|delta slope| within the noise floor).no significant changepoint
changepoint
none detected
snapshot 1 of 1 · · m2
code-leads-citation lag reserved

reserved: citation series too short / axis absent for cross-correlation

status
reserved
value
insufficient data
snapshot 1 of 1 · · m2
axis alpha vs cohort reserved

reserved: cohort N too small for regression (power ~0)

status
reserved
value
insufficient data
snapshot 1 of 1 · · m2
gate ETA reserved

reserved: needs snapshot history (1 snapshot exists)

status
reserved
value
insufficient history
snapshot 1 of 1 · · m2
methodology gate snapshot 1 of 1 · · m2
commit weeks
52
citation years
0
shipped
2
reserved
4

shipped

  • recency-swing
  • commit-changepoint

reserved

  • commit-sharpe
  • code-leads-citation-lag
  • axis-alpha
  • gate-eta
snapshot 1 of 1 · · m2

Adoption trend

SWE-bench dependents over time6 points, range 6 to 7 dependents.
min
6 dependents
max
7 dependents
latest
7 dependents
points
6

Downstream packages depending on pypi/swebench (deps.dev), captured point-in-time each snapshot. An adoption signal, not part of the momentum score.

Verdict

SWE-bench is a mature incumbent included to calibrate its cohort. It is measured but not badge-eligible.

Sources

SWE-bench/SWE-bench development-velocity axis

Embed this badge

Evidaxis calibration, SWE-bench
[![Evidaxis: SWE-bench](https://evidaxis.org/badge/e_EF7QKB37YJK/2026-w32.svg)](https://evidaxis.org/e/e_EF7QKB37YJK)

Others in Coding Agents

Agentless Coding Agents
Single-axis
46
p32
vel -0.31 cit no axis
axis dials
-0.31 n/a
Aider Coding Agents
Single-axis
19
p11
vel -2.50 cit no axis
axis dials
-2.50 n/a
ChatDev Coding Agents
Single-axis
23
p16
vel -2.16 cit no axis
axis dials
-2.16 n/a
Cline Coding Agents
Watch
63
p89
vel +1.00 cit no axis
axis dials
+1.00 n/a
Codex CLI Coding Agents
Single-axis
57
p84
vel +0.57 cit no axis
axis dials
+0.57 n/a
Continue Coding Agents
Single-axis
13
p0
vel -3.00 cit no axis
axis dials
-3.00 n/a

Full Coding Agents cohort →

entity_id e_EF7QKB37YJK methodology m2 snapshot 035a72a81e59 period 2026-w32 record JSON ↓ source snapshot

Score derived from public data on Evidaxis methodology m2, never edited by hand. Cite as: “Evidaxis, SWE-bench, entity e_EF7QKB37YJK, snapshot 2026-w32.”

Durable reference (format-independent): urn:evidaxis:claim:e_EF7QKB37YJK:m2:2026-08-08