Cognition launches FrontierCode 2.0, a maintainer-built mergeability coding benchmark
Cognition introduces FrontierCode 2.0, a coding benchmark built by more than 20 open-source maintainers who each spent over 40 hours per task on realistic evals from repositories they maintain. It measures whether agent output would be mergeable, using grading criteria that include a 'thoroughness' cluster for correctness conditions an agent should infer. Compared with FrontierCode 1.1, tasks draw from 55% more repositories, include 43% more blocking criteria, and reference solutions modify 67% more files. The top model scores 45.5%.
Open original source cognition.comOne primary source; other links are context, not independent corroboration.
- Primary sourceX discovery · automated brief
- 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
Cognition introduces FrontierCode 2.0, a coding benchmark built by more than 20 open-source maintainers who each spent over 40 hours per task on realistic evals from repositories they maintain. It measures whether agent output would be mergeable, using grading criteria that include a 'thoroughness' cluster for correctness conditions an agent should infer. Compared with FrontierCode 1.1, tasks draw from 55% more repositories, include 43% more blocking criteria, and reference solutions modify 67% more files. The top model scores 45.5%.
A READER'S EVIDENCE CHECKHow to read this signalSix questions about claims, sources, risks, and open checks
This is a signal-level guide until claim-level evidence is available. “Not recorded” means the evidence has not been assessed here; it does not mean a result is false. How we check AI news →
- 01Who is making each claim?
The claimant and whether each claim was measured or asserted are not recorded in this public brief.
- 02What was actually tested?
Not recorded per claim. Check the source for sample size, tasks, model and benchmark versions, and reproducibility.
- 03Where are the sources?
Original link: cognition.com. Brief basis: fetched source page. Primary page date and quote locations are not established in this panel.
- 04What could mislead?
Source claims not independently checked; scores and cost figures are as reported by Cognition, and the benchmark is authored by its maker.
- 05What is at risk if wrong?
Not assessed per claim. Consider safety, privacy, cost, and adoption decisions before relying on the result.
- 06What would change the view?
No claim-specific check is recorded. An inspectable primary method or independent replication could change the assessment.
Key findings from the stored brief
- Tasks are drawn from 55% more repositories and include 43% more blocking criteria than FrontierCode 1.1.
- Reference solutions modify 67% more files and 89% more lines of code on average, in codebases 25% larger.
- Thoroughness criteria rose from 168 to 501 blocking criteria, measuring inferred correctness such as fixing every instance of a bug.
- The top-performing model, Claude Opus 5.5, scores 45.5%, which the source describes as far from saturated.
Source claims not independently checked; scores and cost figures are as reported by Cognition, and the benchmark is authored by its maker.
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.
Related archive coverage
These relationships record a citation or topic overlap, not confirmation of the finding.
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.
These are Grok Bot discovery notes. AINewsIQ prepared a brief from available source material. The bot's rationale, linked posts, authors, and stances have not been independently verified.
Cognition's FrontierCode 2.0 grades models on mergeability against rubrics written by open-source maintainers, with larger tasks than v1.1 (67% more files, 89% more lines changed) and new security and performance categories. Opus 5.5 leads at the high end; GPT-6.1 Sol is the cost-efficiency standout. Vendor-run benchmark from a company that sells a coding agent.
Source claims not independently checked; scores and cost figures are as reported by Cognition, and the benchmark is authored by its maker.
Engagement counts, when shown, are observed snapshots. They do not establish accuracy or corroboration.
- x post ↗ · 2,868 likes, 253 reposts, 184 replies, 206 quotes · observed 10/28/2026, 4:11:15 PM
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
A benchmark judging whether agent code would be merged by maintainers targets a gap that conventional test-passing benchmarks miss. It suggests frontier models still fall well short on unstated constraints and design judgment.
Limits and open questions
Source claims not independently checked; scores and cost figures are as reported by Cognition, and the benchmark is authored by its maker.
Organization-specific financial, security, and operational effects require the organization's own context and review.
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.