Termsslug: elo-rating

Elo Rating

Elo 评分

A statistical method calculating relative model rankings based on pairwise comparison results, commonly used in arena-style evaluations.

Curated Intelligence

4 items · Page 1/1
Artificial Analysis (@ArtificialAnlys)T1·Agents
45 pts
Cartesia Voice Agents Score Higher in Arena Benchmark

Cartesia Managed Agents achieved an Elo score of 1115 on the Artificial Analysis Speech Agent Arena, a 98-point improvement over its predecessor. Task success rates rose from 77.5% to 91.7%, driven by a cascaded architecture combining Ink STT/TTS models with GPT-5.5.

机器之心 (微信公众号)T2·Models
68 pts
Film Studio Utopai Ranks #2 in Video Arena via Production System

Utopai Studios' custom model Utopai X ranked second globally in Artificial Analysis's video blind test. The breakthrough lies in embedding generation capabilities into a complete film production workflow, demonstrating that an 'intelligent production system' outperforms pure parameter scaling for long-narrative video.

Artificial AnalysisT1·Models
86 pts
Claude Opus 5.5 Tops Artificial Analysis Index with Record 58 Score and Price Cuts

Anthropic's Claude Opus 5.5 achieves a record-high score of 58 on the Artificial Analysis Intelligence Index at max effort. It matches GPT-6 Astra on key benchmarks like Terminal-Bench 4.0 while leading in six out of ten evaluations, particularly excelling in agentic knowledge work (AA-Briefcase). The release also includes significant price reductions to $4/$20 per million tokens.

Artificial AnalysisT1·Models
68 pts
Grok 4.7 Benchmark: Intelligence Index Hits 46, Coding Agent Performance Overtakes GPT-5.6 Sol

xAI's Grok 4.7 scores 46 on the Artificial Analysis Intelligence Index, placing it among the top four AI labs. The model shows significant gains in long-horizon knowledge work and coding agent tasks, with a Coding Agent Index score of 56 that surpasses GPT-5.6 Sol. While improvements in non-agentic areas are incremental, these specific breakthroughs establish a new competitive dynamic.