01
Click endurance
29k
saved
42.7%
less token use
91k baseline to 62k with Kairn.
10-turn debugging session.
Evidence
Early evidence from source-heavy coding work.
Controlled runs show Kairn reducing token use, keeping long sessions focused, and improving source routing. The detailed split is available for deeper review.
These results measure context governance during coding work, not raw search quality.
Token use
Tokens saved after quality passes.
These rows compare successful runs against successful runs. Bars show percent reduction; labels show tokens saved.
Evidence sets
8+
headline sets plus latest check
Median reduction
42.7%
clean rows shown
Official tasks passed
20/25
Kairn across official campaigns
Endurance suites
4
long-session evidence
Latest fresh check
Fresh source-routing check
3 fresh pass/pass rows saved 48k tokens in aggregate: 123k baseline to 75k with Kairn. This is repeat-1 evidence, so it is tracked separately from the headline examples.
38.7%
aggregate reduction
Fresh official slice
Fresh SWE-Pro governance slice
Kairn passed 9/9 valid official passes in the latest branch check, with 0 harmful assists and 0 token regressions. Clean pass/pass rows still saved tokens, but the stronger signal was source and scope discipline.
9.3%
clean median
02
Node SemVer endurance
11k
saved
16.4%
less token use
70k baseline to 58k with Kairn.
Fresh long-session JavaScript package test.
03
OpenLibrary SWE-Pro
12k
saved
54.2%
less token use
22k baseline to 10k with Kairn.
Official evaluator row.
Coverage
More than a token chart.
Kairn has been tested across clean token-savings rows, longer sessions, MCP delivery, official evaluator rows, and governor ablations.
Token reduction
On clean rows, Kairn reduced token use by 16.4-54.2% after both sides completed the task.
measured
Long sessions
Endurance runs show Kairn keeping source focus across repeated debugging and verification turns.
repeat-tested
Editor path
MCP-first runs recorded real tool calls and useful source guidance, so Kairn is not limited to terminal hooks.
MCP
Official evaluator
Across 25 official SWE-Pro rows from evidence campaigns, Kairn passed 20/25 versus 15/25 baseline.
SWE-Pro
Fresh official slice
In the latest 10-row SWE-Pro governance slice, Kairn passed 9/9 valid official evaluations with zero harmful assists.
fresh
Quality lift
Several controlled rows passed with Kairn when the baseline missed quality, showing the value of better source focus.
rescue
Decision layer
Ablation showed the governor improves safety by shrinking, suppressing, or asking for evidence instead of over-assisting.
governor
Latest fresh check
A post-launch source-routing check saved 47,641 tokens across 3 fresh pass/pass rows; repeat-3 is still needed before treating it as headline evidence.
fresh
Quality
Better source focus can improve outcomes.
Some controlled rows passed with Kairn after the baseline missed quality. Those rows are shown separately from token-savings rows.
official evaluator pass rate
Official SWE-Pro campaigns
25 official rows across evidence campaigns; clean savings and quality/source-focus rows are separated.
Codex alone
15/25
tasks passed
60%
pass rate
Codex + Kairn
20/25
tasks passed
80%
pass rate
Interpretation
What this does and does not prove.
Kairn is strongest when the task would otherwise cause repeated source searching.
Kairn's silence is a feature: it avoids adding context when context is not useful.
The strongest savings rows count only after quality gates pass.
Codex session is the most-tested path; MCP is the portable editor path.
Details
Want the benchmark split?
The detailed page keeps modes, MCP runs, official evaluator rows, and caveats visible for technical review.