benchmark · model-level

AA Coding Index

Coding Index·27 models·homepage ↗

Rank · benchmarks#16 / 19

Score29

How it works

Artificial Analysis's coding-specific composite — a blend of coding evals (SWE-bench, Terminal-Bench, SciCode, LiveCodeBench and others) into one index per model.

Metric

Coding Index

Models

Type

Model-level

How we ranked it

composite 29 / 100

Every benchmark is scored on four weighted criteria — how directly it measures a real coding agent, how much of the stack it covers, how real its tasks are, and how open and reproducible it is. Those blend into the composite that ranks it #16 of 19.

Agent-nativeweight 30%

Scores a real (harness × model) pair — the full agent — not just the bare model.

Stack coverageweight 30%

Data-driven: this benchmark exposes 27 models.

Task realismweight 25%

Executable, real-world coding tasks with hard pass/fail — not human preference or aggregate scores.

Open & reproducibleweight 15%

Open data with per-run receipts you can audit.

Our review

More on-target than the general Intelligence Index because it's coding-only, and a reasonable quick proxy. But it's a composite of other benchmarks (some of which we already track directly), so it double-counts, and it's still model-level rather than agent-level.