Model result · rank #7

Gemini 3.5 Flash

OpenRouter · google/gemini-3.5-flash. Public result card with the model’s overall score, lane measurements, runtime/cost telemetry, and ranking formula.

Download source JSON Compare all models

Overall score81.19

Rank #7

Full / Agentic88.79

Full rank #3

SWE MVP73.59

SWE rank #8

Measured cost$1.502

100.0% reliability

Overall

All-around publication view

Score81.19

Formula50% Full + 50% SWE

BasisOpenRouter · google/gemini-3.5-flash

The overall score is calculated from the Full/Agentic and SWE lanes, keeping the aggregate comparable while preserving the measurements behind it.

Lane 01

Full / Agentic benchmark

Final88.79

Capability93.55

Agentic91.87

Pass rate93.0%

Prompts43

This lane captures instruction following, structured behavior, tool discipline, and general agentic reliability.

Lane 02

Software engineering MVP

SWE score73.59

Focused final72.96

Capability70.23

Daily driver73.50

Prompts24

This lane is closer to implementation usefulness: source handling, architecture cleanliness, and deliverable quality.

Telemetry

Runtime economics

Full cost$1.011

SWE cost$0.491

Full avg seconds7.23

SWE time290.68s

Decode99.62

Cost, time, and runtime basis are telemetry. They explain tradeoffs; they do not secretly overwrite the capability scores.

Interpretation

Why this result lands here.

The model is stronger in the Full/Agentic lane than in the SWE lane; the overall score is therefore shown with both component lanes visible.