Electricity Bench

cursor-cli + composer-2.5

cursor-cli harness · Cursor Pro ($20/mo)

DOVERALL

Clears tier 2 of 5 (partial at tiers 3 and 5) on Real-world issues

Capability mean across 3 graded suite families.

capability
57%
plan cost per run
1.8% of Cursor Pro
median task time
1m 30s
reliability
100%
12345
What the tiers mean

"Clears" marks the highest rung where every task at it and below was solved. Tier 1: localized single-file bug · tier 3: cause in a different module than the symptom · tier 5: problems that took a human hours. A cleared rung solved every task on it; a partial rung solved some. The headline stops at the last unbroken rung, so a solved tier above a half-solved one counts only as partial.

Clears level 1 of 5 (partial at levels 2, 3, 4 and 5) on Vibe coding v1 - the vaguest report level at which it still fixed every bug, having also cleared every richer level below
12345
What the levels mean

Level 1: a full brief naming the surface and the acceptance · level 3: the report as filed · level 5: a vague vibe report in a non-technical user's words. Context falls as the level rises, so vaguer is harder. A cleared level fixed every bug at it and below; a partial level fixed some. The headline stops at the last unbroken level, so a fix above a half-cleared level counts only as partial.

5 of this suite's tasks were credited rather than run: on a suite that asks the same problem at several levels of detail, solving it from the vaguest description credits the more detailed ones instead of asking again.

Per-suite results

C
Real-world issues v3
57%
plan cost per run
1.8% of Cursor Pro
median task time
1m 30s
reliability
100%
tier 1tier 2tier 3tier 4tier 5

11 of 15 tasks solved - grouped by tier, easiest first; hover a square for its time and turns.

Run on cursor-cli 2026.08.11-e8db854, model composer-2.5, week 2026-W33.

One run of this suite consumes 1.8% of Cursor Pro.

As work per window: about 57 suite runs per month on this plan - 2.84 runs per plan-dollar (at $20/mo, priced 2026-07-20). Work figures compare outputs and are comparable across agents; the raw quota share above is not.

How the plan cost was measured

Measured per suite run (estimated-from-telemetry); an absolute cost on this plan, not comparable with another agent's plan. Measured in a declared quiet window on a shared subscription: operator Cursor Pro login (hericko.aljaz@gmail.com) shared between eb-runner and the claude host; graded runs only inside the scheduled quiet window, all recurring tasks blocked for its duration.

D
Spec planning v1
51%
plan cost per run
0.6% of Cursor Pro
median task time
4m 30s
reliability
100%

0 of 4 tasks solved - hover a square for its time and turns.

Run on cursor-cli 2026.08.11-e8db854, model composer-2.5, week 2026-W34.

One run of this suite consumes 0.6% of Cursor Pro.

As work per window: about 173 suite runs per month on this plan - 8.66 runs per plan-dollar (at $20/mo, priced 2026-07-20). Work figures compare outputs and are comparable across agents; the raw quota share above is not.

How the plan cost was measured

Measured per suite run (estimated-from-telemetry); an absolute cost on this plan, not comparable with another agent's plan. Measured in a declared quiet window on a shared subscription: operator Cursor Pro login (hericko.aljaz@gmail.com) shared between eb-runner and the claude host; graded runs only inside the scheduled quiet window, all recurring tasks blocked for its duration.

D
Vibe coding v1
37%
plan cost per solve
0.1% of Cursor Pro
median task time
1m 51s
reliability
100%
level 5level 4level 3level 2level 1issue aissue bissue cissue dissue e

15 of 25 tasks solved (5 credited without running) - each row is one bug, vaguest report first and context growing to the right; hover a square for its time and turns.

Run on cursor-cli 2026.08.11-e8db854, model composer-2.5, week 2026-W34.

One good solve on this suite consumes 0.1% of Cursor Pro on average.

As work per window: about 822 good solves per month on this plan - 41.11 solves per plan-dollar (at $20/mo, priced 2026-07-20). Work figures compare outputs and are comparable across agents; the raw quota share above is not.

How the plan cost was measured

Counted over each issue's first solve (5 solves); the walk's failed harder cells and confirmation runs are the benchmark's own search cost and are excluded (full walk: 2.8% over 20 cells). Measured per cell run (estimated-from-telemetry); an absolute cost on this plan, not comparable with another agent's plan. Measured in a declared quiet window on a shared subscription: operator Cursor Pro login (hericko.aljaz@gmail.com) shared between eb-runner and the claude host; graded runs only inside the scheduled quiet window, all recurring tasks blocked for its duration.

Task outputs from these runs are withheld: publishing them would reveal the private suite. Per-task scores, times and turns are in the chips above.

Trajectory

One point per grading week - a re-run within a week shows only its latest execution. Grades compare only within a suite version, so each version gets its own line; while a suite is being retired both versions run, and the superseded one is dashed. The filled point is the run the scorecard shows today. Click a point or week for plan usage. All agents over time →

Suite
Real-world issues · 3 weeks graded · latest 2026-W34
40%50%60%70%80%2026-W322026-W332026-W34cursor-cli + composer-2.5 · 2026-W32 · Aug 4, 2026 · 57% · Ccursor-cli + composer-2.5 · 2026-W33 · Aug 10, 2026 · 60% · Ccursor-cli + composer-2.5 · 2026-W34 · Aug 16, 2026 · 57% · C

Weekly heads

Newest week first. Click a row for usage.

WeekGradeCapabilityClearsTask timeUsage
2026-W34 · v3nowC57%tier 2 · p51m 30s1.8% of Cursor Pro
2026-W33 · v3C60%tier 4 · p53m 52s1.9% of Cursor Pro
2026-W32 · v3C57%tier 2 · p51m 57s1.7% of Cursor Pro

Frequently asked questions

Is composer-2.5 worth it on Cursor Pro?

On our private real-world suite (real-world@v3), composer-2.5 earned grade C with 57% capability, clearing every problem up to tier 2 of 5 - a bug spanning a few files and landing some tier 5 problems. It runs on Cursor Pro at $20/month (public price as of 2026-07-20). One full suite run consumed 1.8% of the plan's month window. Measured 2026-08-16.

How good is composer-2.5 at real-world coding?

composer-2.5 clears tier 2 of our five-tier real-world ladder - every task at that tier and below - where tier 1 is a localized single-file bug and tier 5 is a problem that took a human engineer hours. It solves some but not all tasks up at tier 5. Its capability score on real-world@v3 is 57% (grade C). Last benchmarked 2026-08-16.

How much coding work does Cursor Pro buy with composer-2.5?

Roughly 57 suite runs per month on Cursor Pro, or 2.84 runs per plan-dollar at $20/month (priced 2026-07-20). Usage is read from the provider's own meter per suite run; the raw quota share is an absolute figure on this plan, not comparable with another agent's plan.

Compare