Electricity Bench

claude-code + claude-haiku-4-5

claude-code harness · Claude Max 5x

plan
Claude Max 5x
price
$100/mo (2026-07-20)
allowance
100 provider-percent per 5h
isolation
quiet-window
usage method
reported-delta

operator Claude Max 5x account on eb-runner, shared with the operator IDE panes; quiet-window until a dedicated benchmark login exists

Solves up to tier 5 on real-world@v3 - the highest difficulty rung with a solved task
12345

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.

overall pending Two distinct suite families are required for an overall grade.

Per-suite results

F
real-world@v3
30%
plan cost per run
17.0% of Claude Max 5x
latency med / p95
188763 ms / 374887 ms
reliability
100%
error rate
0%

Run on claude-code 2.1.221 (Claude Code), model claude-haiku-4-5 (--effort medium), Aug 4, 2026.

One run of this suite consumes 17.0% of Claude Max 5x - about 5.9 runs per 5h. Measured per suite run (reported-delta); an absolute cost on this plan, not comparable with another agent's plan. Measured in a declared quiet window on a shared subscription: operator Claude Max 5x account on eb-runner, shared with the operator IDE panes; quiet-window until a dedicated benchmark login exists.

As work per window: about 859 suite runs per month on this plan - 8.59 runs per plan-dollar (at $100/mo, priced 2026-07-20). Work figures compare outputs and are comparable across agents; the raw quota share above is not.

Trajectory

Every scored run of this agent, oldest first. 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within a suite version, so each version gets its own line; while a suite is being retired both versions run, and the superseded one is dashed.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. All agents over time →

real-world · 1 run · latest Aug 4, 2026
Capability
0%50%100%Aug 4claude-code + claude-haiku-4-5 · v3 · Aug 4, 2026 · 30% · grade F0%50%100%Aug 4claude-code + claude-haiku-4-5 · v3 · Aug 4, 2026 · 30% · grade F
Median latency
151010 ms188763 ms226516 msAug 4claude-code + claude-haiku-4-5 · v3 · Aug 4, 2026 · 188763 ms · grade F151010 ms188763 ms226516 msAug 4claude-code + claude-haiku-4-5 · v3 · Aug 4, 2026 · 188763 ms · grade F
Plan cost per run - a share of this agent's own plan, not a cross-agent number
13.6%17.0%20.4%Aug 4claude-code + claude-haiku-4-5 · v3 · Aug 4, 2026 · 17.0% · grade F13.6%17.0%20.4%Aug 4claude-code + claude-haiku-4-5 · v3 · Aug 4, 2026 · 17.0% · grade F

Aug 4, 2026: F on v3

Compare

Regrade changelog

Transcript excerpts

Outputs are shown only when they can be separated safely from the private suite. Scores and run metadata remain visible when an output is withheld.

real-world@v3
Task#TurnsLatencyOutput
b2r5h8134144664 mswithheldPrivate-suite content
c8g2m4126151818 mswithheldPrivate-suite content
d9b4x6129139077 mswithheldPrivate-suite content
h5n2v7136188763 mswithheldPrivate-suite content
h7w3j5129110449 mswithheldPrivate-suite content
k4f9t7167488091 mswithheldPrivate-suite content
m6t3q8151209331 mswithheldPrivate-suite content
p1x9k4139229902 mswithheldPrivate-suite content
q8k3j6126156793 mswithheldPrivate-suite content
r3v8m5149219782 mswithheldPrivate-suite content
s2j7f4143229761 mswithheldPrivate-suite content
t5s8n2131130023 mswithheldPrivate-suite content
w4j7q2142265089 mswithheldPrivate-suite content
y3p7k1138179931 mswithheldPrivate-suite content
z6q1v9161326372 mswithheldPrivate-suite content