이미지: X — 벤치마크·평가
Summary
- Meta has released Muse Spark 1.2, which scored 54 on the Intelligence Index tracked by evaluator Artificial Analysis.
- The model marks Meta's third release in four months and was noted for significant improvement in agentic knowledge work capabilities compared to the previous version.
- The score of 54 reportedly ties the model for third place alongside SpaceXAI, though detailed category-by-category scores were not disclosed.
- 모델명
- Muse Spark 1.2
- 개발사
- 메타(Meta)
- 평가 지표
- Artificial Analysis Intelligence Index
- 점수
- 54점
- 순위
- SpaceXAI와 공동 3위
- 릴리스 빈도
- 4개월 동안 세 번째 공개
- 개선 영역
- 에이전틱 지식 노동(agentic knowledge work) 능력
- 공개 확인 시점
- 2026년 8월 5일(평가 기관 게시 기준)
Meta releases Muse Spark 1.2
Meta has released a new model called Muse Spark 1.2. Evaluator Artificial Analysis reported that the model scored 54 on its Intelligence Index. This score matches that of SpaceXAI, putting the two in a tie for third place, according to the evaluator. The disclosure did not specify which models hold the first and second spots, nor the exact score gaps between models.
Third release in four months
One notable point is the pace of release. According to Artificial Analysis, Muse Spark 1.2 is Meta's third model release within a four-month span. The pattern of frequent version updates aimed at raising scores appears to reflect an effort to close the gap with top-tier competitors.
| Item | Confirmed detail |
|---|---|
| Model | Muse Spark 1.2 |
| Intelligence Index | 54 |
| Ranking | Tied for third (with SpaceXAI) |
| Release cadence | 3 releases in 4 months |
| Key improvement | Agentic knowledge work |
Improvement focus: "agentic knowledge work"
The evaluator noted a clear improvement in agentic task performance compared to the previous release. Artificial Analysis described it as "significantly improving agentic knowledge work capabilities." This suggests gains in multi-step tasks involving documents, search, and tool use, rather than single-turn question answering. However, no detailed figures were provided on which specific sub-benchmarks improved or by how much.
For ongoing coverage of Intelligence Index scores across models and shifts in agent performance evaluation methods, see METAL LAB's benchmark coverage.
What remains unconfirmed
The information currently available is based on a single post from the evaluator. Technical specifications — including whether model weights are open, context length, pricing, and training/inference infrastructure — remain unconfirmed. The specific model name for SpaceXAI, mentioned as tying at third place, was also not specified.
| Category | Status |
|---|---|
| Intelligence Index score | Confirmed |
| Ranking | Confirmed (tied for third) |
| Detailed benchmark breakdown | Undisclosed |
| Model specs/release format | Undisclosed |
| Top 1st/2nd place models | Not mentioned |
Further benchmark results and an official announcement from Meta are expected to clarify the significance of the score and the competitive landscape.
