PulseAugur
EN
LIVE 01:33:14

Replit launches ViBench 2.0 for long-horizon AI software development

By PulseAugur Editorial · [1 sources] ·

Replit has launched ViBench 2.0, a new benchmark designed to evaluate AI models on long-horizon software development tasks. Unlike previous benchmarks that focus on specific coding problems, ViBench 2.0 requires AI agents to build entire products from scratch, adding features incrementally over time. The benchmark includes seventeen distinct products and 241 feature requests, with top-performing models like GPT-6.1 Sol, GPT-6 Astra, and Opus 5.5 achieving success rates between 49-65% on the test plans. AI

IMPACT Establishes a new standard for evaluating AI's ability to handle complex, multi-stage software development projects.

RANK_REASON New benchmark for evaluating AI models on complex software development tasks. [lever_c_demoted from research: ic=1 ai=1.0]

Read on Replit blog →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Replit launches ViBench 2.0 for long-horizon AI software development

How we ranked this

Signal score
9 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
New benchmark for evaluating AI models on complex software development tasks. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
product, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Same-day
Cluster formed today. Ranking reflects the current source set at time of score.

Full methodology in our editorial standards.

COVERAGE [1]

  1. Replit blog TIER_1 English(EN) ·

    ViBench 2.0: A Benchmark for Long-Horizon Software Building

    Today we’re releasing ViBench 2.0, a benchmark for long-horizon software development. It joins the Specialized Intelligence Index (SII) from Fireworks [1], our partner for open-weight model inference and training. The SII brings together practitioner-built benchmarks to compare o…

RELATED ENTITIES

RELATED TOPICS