--- title: "hermes \u00b7 Gemini 3.5 Flash | Pelican benchmark" url: https://evaljetty.com/pelicans/runs/agents/hermes-flash.html description: "hermes \u00b7 Gemini 3.5 Flash (openrouter/google/gemini-3.5-flash) drawing a pelican riding a bicycle: every hill-climb run with self-scores, external judge scores and Jetty trajectories." updated: 2026-09-28 publisher: Jetty (https://jetty.io) --- 02 — Runs · v1 · jettyio · auto-evolved runbook (weakest-axis targeting) # hermes openrouter/google/gemini-3.5-flash Same hermes agent, Sonnet swapped for Flash: +5 self-score points from the model alone. Judge, best round 28.2/40 round v9 (agent's pick) Self-score 39.5/40 inflation +11.3 Runs 10 not recorded (v1) ![hermes · Gemini 3.5 Flash best round](https://evaljetty.com/pelicans/img/hermes-flash/v9.svg) best round v9 · e6d4548b ### Self-score vs judge, by axis Best round v9, 0–10 per axis agent self-scoreexternal judge **Judge:** The bicycle is well formed, with spoked wheels, chain, crank, fenders and a fish crate, set against an attractive seaside backdrop. The pelican's anatomy is confused: the huge orange pouch reads as a detached sheet, the head is tiny and tilted back, the body is a jumble of white blobs with an odd grey tail, and the feet only loosely meet the pedals. ### 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 ## 10 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/index.html#all-runs). [![hermes · Gemini 3.5 Flash round v1](https://evaljetty.com/pelicans/img/hermes-flash/v1.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#hermes-flash/v1) **v1** passjudge **28.5** · self 39 c118fb80[runbook](https://evaljetty.com/pelicans/author/versions.html#hermes-flash/v1) [![hermes · Gemini 3.5 Flash round v2](https://evaljetty.com/pelicans/img/hermes-flash/v2.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#hermes-flash/v2) **v2** passjudge **28.2** · self 38 2128ffc1[runbook](https://evaljetty.com/pelicans/author/versions.html#hermes-flash/v2) [![hermes · Gemini 3.5 Flash round v3](https://evaljetty.com/pelicans/img/hermes-flash/v3.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#hermes-flash/v3) **v3** failjudge **26.5** · self 37 2405db86[runbook](https://evaljetty.com/pelicans/author/versions.html#hermes-flash/v3) [![hermes · Gemini 3.5 Flash round v4](https://evaljetty.com/pelicans/img/hermes-flash/v4.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#hermes-flash/v4) **v4** passjudge **28.7** · self 39 822a3bca[runbook](https://evaljetty.com/pelicans/author/versions.html#hermes-flash/v4) [![hermes · Gemini 3.5 Flash round v5](https://evaljetty.com/pelicans/img/hermes-flash/v5.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#hermes-flash/v5) **v5** passjudge **30** · self 38.8 60502506[runbook](https://evaljetty.com/pelicans/author/versions.html#hermes-flash/v5) [![hermes · Gemini 3.5 Flash round v6](https://evaljetty.com/pelicans/img/hermes-flash/v6.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#hermes-flash/v6) **v6** failjudge **26.8** · self 38 2c2a74e8[runbook](https://evaljetty.com/pelicans/author/versions.html#hermes-flash/v6) [![hermes · Gemini 3.5 Flash round v7](https://evaljetty.com/pelicans/img/hermes-flash/v7.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#hermes-flash/v7) **v7** failjudge **27** · self 38.5 c165de92[runbook](https://evaljetty.com/pelicans/author/versions.html#hermes-flash/v7) [![hermes · Gemini 3.5 Flash round v8](https://evaljetty.com/pelicans/img/hermes-flash/v8.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#hermes-flash/v8) **v8** failjudge **26.3** · self 38 d391e018[runbook](https://evaljetty.com/pelicans/author/versions.html#hermes-flash/v8) [![hermes · Gemini 3.5 Flash round v9](https://evaljetty.com/pelicans/img/hermes-flash/v9.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#hermes-flash/v9) **v9** · best passjudge **28.2** · self 39.5 e6d4548b[runbook](https://evaljetty.com/pelicans/author/versions.html#hermes-flash/v9) [![hermes · Gemini 3.5 Flash round v10](https://evaljetty.com/pelicans/img/hermes-flash/v10.svg)](https://evaljetty.com/pelicans/investigations/best-pelican.html#hermes-flash/v10) **v10** failjudge **26.5** · self 38.5 3e21bf05[runbook](https://evaljetty.com/pelicans/author/versions.html#hermes-flash/v10) ### Scores per round | Round | Pelican | Bicycle | Composition | Polish | Self total | Judge total | Verdict | Judge spread | Cost | Minutes | Trajectory | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | v1 | 7self 10 | 7.5self 9.5 | 6.8self 10 | 7.2self 9.5 | 39 | 28.5 | pass | 1 | – | – | c118fb80 | | v2 | 7self 9 | 7.8self 10 | 6.3self 9 | 7self 10 | 38 | 28.2 | pass | 1 | – | – | 2128ffc1 | | v3 | 6.3self 9.5 | 7.3self 9 | 6.3self 9.5 | 6.5self 9.5 | 37 | 26.5 | fail | 0 | – | – | 2405db86 | | v4 | 7self 10 | 7.7self 10 | 6.8self 9.5 | 7.2self 9.5 | 39 | 28.7 | pass | 1.5 | – | – | 822a3bca | | v5 | 7self 9.5 | 8.3self 9.5 | 7.2self 10 | 7.5self 9.8 | 38.8 | 30 | pass | 0 | – | – | 60502506 | | v6 | 6.3self 9.5 | 7.5self 9.5 | 6.3self 9.5 | 6.7self 9.5 | 38 | 26.8 | fail | 0.5 | – | – | 2c2a74e8 | | v7 | 5.8self 9.8 | 7.8self 9.5 | 6.7self 9.5 | 6.7self 9.7 | 38.5 | 27 | fail | 2.5 | – | – | c165de92 | | v8 | 5.5self 9 | 8self 10 | 6.2self 9 | 6.7self 10 | 38 | 26.3 | fail | 1 | – | – | d391e018 | | v9 | 6self 9.5 | 8self 10 | 6.8self 10 | 7.3self 10 | 39.5 | 28.2 | pass | 2 | – | – | e6d4548b | | v10 | 5.3self 9.8 | 8self 9.5 | 6.3self 9.5 | 6.8self 9.7 | 38.5 | 26.5 | fail | 1 | – | – | 3e21bf05 | --- Source: https://evaljetty.com/pelicans/runs/agents/hermes-flash.html · Evals by Jetty · Run your own evals: https://jetty.io/?utm_source=evaljetty&utm_medium=llms