Artificial Analysis launches Coding Agent Index v2.0, swapping DeepSWE for Cognition FrontierCode v1.1
Artificial Analysis describes its Coding Agent Index v2.0, an equal-weight average of three benchmarks: FrontierCode v1.1, Terminal-Bench 4.0, and SWE-Atlas-QnA-AA. The methodology covers 290 tasks and scores coding agents on end-to-end software engineering work alongside reliability, token usage, cost, and execution time. Scoring uses pass@1 for two benchmarks and mean rubric-based scores for FrontierCode. Attempts that error, time out, or hit a safety refusal score zero, and refusals are detected with deterministic rules and an LLM judge.
One primary source; other links are context, not independent corroboration.
- Primary sourceX discovery · concern recorded
- Discovery linkX
Grok Bot submitted the link. AINewsIQ fetched the primary source and prepared a source-based brief automatically. An editor has not reviewed the story; social attention and the brief do not independently verify its claims.
Open full article view →- Linked article+30
- Fetched source page+30
- No editor link review+0
- Original non-discussion link+5
This measures the stored source trail and editorial steps. It does not certify accuracy, peer review, or independent corroboration. Popularity adds no points.
View full scoring →Read the scoring rubric →What the source says
Artificial Analysis describes its Coding Agent Index v2.0, an equal-weight average of three benchmarks: FrontierCode v1.1, Terminal-Bench 4.0, and SWE-Atlas-QnA-AA. The methodology covers 290 tasks and scores coding agents on end-to-end software engineering work alongside reliability, token usage, cost, and execution time. Scoring uses pass@1 for two benchmarks and mean rubric-based scores for FrontierCode. Attempts that error, time out, or hit a safety refusal score zero, and refusals are detected with deterministic rules and an LLM judge.
Key findings from the stored brief
- Coding Agent Index v2.0 is the equal-weight average of FrontierCode v1.1, Terminal-Bench 4.0 and SWE-Atlas-QnA-AA.
- The index covers 290 tasks: 100 FrontierCode, 66 Terminal-Bench 4.0 and 124 SWE-Atlas-QnA-AA.
- FrontierCode attempts fail a blocking check score zero, and attempts that fetch the upstream solution also score zero.
- Attempts that error, exceed time limits or are blocked by a safety refusal score zero on every benchmark.
Source claims not independently checked; the methodology describes Artificial Analysis's own benchmark construction and judge configuration.
This is a source-based abstract, not independent validation of the source's claims.
Evidence & sources
These links show where the signal came from and what the source page linked to. A linked page is context, not independent corroboration.
Grok Bot submitted the link. AINewsIQ fetched the primary source and prepared a source-based brief automatically. An editor has not reviewed the story; social attention and the brief do not independently verify its claims.
Timeline & updates
- Signal dated by source or discovery feed
This timestamp places the signal in the radar. It may be the discovery post date rather than the primary page's first publication date.
- Shared brief stored
AINewsIQ prepared one abstract from fetched source page.
A publication revision history is not stored for this signal. New engagement snapshots do not mean the source itself changed.
Why this may matter
The page documents how a widely cited coding-agent leaderboard is built, including grading rules, anti-cheating checks and how safety refusals are counted. Readers comparing agent results need these details to interpret scores.
Limits and open questions
Source claims not independently checked; the methodology describes Artificial Analysis's own benchmark construction and judge configuration.
Organization-specific financial, security, and operational effects require the organization's own context and review.
Discussion
Ask about the source, add context, or point readers to a relevant follow-up.
Discussion
Comments are reviewed before publication. Discuss the evidence, not the person.
No published comments yet.
Reader discussion opens with accounts. You can read the original source and published comments now.