---
title: "claude-code \u00b7 Haiku 5.5 | Pelican benchmark"
url: https://evaljetty.com/pelicans/runs/agents/v2-claude-haiku55.html
description: "claude-code \u00b7 Haiku 5.5 (anthropic/claude-haiku-5.5) drawing a pelican riding a bicycle: every hill-climb run with self-scores, external judge scores and Jetty trajectories."
updated: 2026-10-07
publisher: Jetty (https://jetty.io)
---

02 — Runs · v2 · jettygrowthteam · auto-evolved runbook (weakest-axis targeting)

# claude-code anthropic/claude-haiku-5.5

claude-code driving anthropic/claude-haiku-5.5 through OpenRouter, 5 hill-climb rounds from the v1 runbook.

Judge, best round

28/40

round v1 (agent's pick)

Self-score

29/40

inflation +1

Runs

5

$0.14 metered

![claude-code · Haiku 5.5 best round](https://evaljetty.com/pelicans/img/v2-claude-haiku55/v1.svg)

best round v1 · 10ad0a5a

### Self-score vs judge, by axis

Best round v1, 0–10 per axis

agent self-scoreexternal judge

**Judge:** The pelican reads clearly: a long orange beak, a large pink pouch, a white body, and a gray wing stretched toward the handlebars, though the neck and body shapes are somewhat crude. The bicycle has spoked wheels, a frame, a seat, a crank and a chain, but the frame geometry is odd: the rear stays sit oddly low, the fork is thin, and the handlebars attach awkwardly. The bird sits on the seat and its legs reach the pedals convincingly, and the clean flat colors give a pleasant overall look.

### Score by round

Out of 40. Each round starts from the previous round's best SVG. Dashed line: the 28/40 pass threshold.

self-scorejudgepass ≥ 28

## 5 drawings, *oldest first.*

Click a drawing to open it in the head-to-head viewer. The outlined card is the agent's best round. Verdicts are derived (code checks on the stored SVG and judge ≥ 28/40); see [Runs](https://evaljetty.com/pelicans/runs/#all-runs).

[![claude-code · Haiku 5.5 round v1](https://evaljetty.com/pelicans/img/v2-claude-haiku55/v1.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#v2-claude-haiku55/v1)

**v1** · best failjudge **28** · self 29
10ad0a5a

[![claude-code · Haiku 5.5 round v2](https://evaljetty.com/pelicans/img/v2-claude-haiku55/v2.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#v2-claude-haiku55/v2)

**v2** passjudge **28.8** · self 29
88a41720

[![claude-code · Haiku 5.5 round v3](https://evaljetty.com/pelicans/img/v2-claude-haiku55/v3.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#v2-claude-haiku55/v3)

**v3** passjudge **28.2** · self 29
2bb59acd

[![claude-code · Haiku 5.5 round v4](https://evaljetty.com/pelicans/img/v2-claude-haiku55/v4.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#v2-claude-haiku55/v4)

**v4** failjudge **26.2** · self 29
6dbc05e0

[![claude-code · Haiku 5.5 round v5](https://evaljetty.com/pelicans/img/v2-claude-haiku55/v5.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#v2-claude-haiku55/v5)

**v5** failjudge **26.2** · self 27
eefa700a

### Scores per round

| Round | Pelican | Bicycle | Composition | Polish | Self total | Judge total | Verdict | Judge spread | Cost | Minutes | Trajectory |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| v1 | 7self 7 | 6.8self 8 | 7.3self 7 | 6.8self 7 | 29 | 28 | fail | 1.5 | $0.03\* | 2.3 | 10ad0a5a |
| v2 | 7.5self 7 | 7self 8 | 7.3self 7 | 7self 7 | 29 | 28.8 | pass | 0.5 | $0.04\* | 1.7 | 88a41720 |
| v3 | 7.2self 7 | 7self 8 | 7.2self 7 | 6.8self 7 | 29 | 28.2 | pass | 2.5 | $0.02\* | 1.4 | 2bb59acd |
| v4 | 7self 8 | 6self 8 | 6.7self 6 | 6.5self 7 | 29 | 26.2 | fail | 0.5 | $0.02\* | 1.3 | 6dbc05e0 |
| v5 | 6.8self 7 | 6.5self 8 | 6.3self 6 | 6.5self 6 | 27 | 26.2 | fail | 1 | $0.02\* | 1.3 | eefa700a |

---
Source: https://evaljetty.com/pelicans/runs/agents/v2-claude-haiku55.html · Evals by Jetty · Run your own evals: https://jetty.io/?utm_source=evaljetty&utm_medium=referral&utm_campaign=evaljetty-launch-2026-09&utm_content=markdown-footer-run-your-own