Geekbench 7
geekbench.com
July 23, 2026
3 min read
🔥🔥🔥🔥🔥
46/100
Summary
Geekbench 7 features new and improved workloads for measuring CPU and GPU performance, including media workloads for audio and video encoding, decoding, and processing. It is available for download on Android, iOS, Windows, macOS, and Linux.
Key Takeaways
- Geekbench 7 introduces new media workloads for measuring CPU performance in audio and video encoding, including tasks like screen-sharing video encoding and real-time caption generation.
- The multi-core benchmark in Geekbench 7 has been redesigned to accurately reflect real-world application behavior, only running tasks in multi-threaded mode when applicable.
- The GPU benchmark now focuses on machine learning and content creation applications, adding workloads for face tracking, image upscaling, and background blurring in video conferencing.
- Geekbench 7 features updated data sets for its workloads to better reflect the larger and more complex files users work with today.
Coverage & discussions
Hacker News · geekbench.com
Community discussions
Hacker News discussion ↗
Historical count: 5 · 5 saved in sample · Historical sample
Analysis ready
Community Sentiment
Positives
- The diverse set of benchmarks in Geekbench allows for comparisons across different platforms, which some find valuable for assessing overall performance.
- Commenters appreciate the sub-scores for different application types in Geekbench, suggesting it provides a broad measure of processor speed.
Concerns
- Many find Geekbench's synthetic benchmarks to be borderline useless, arguing that a higher score doesn't translate to real-world performance differences.
- Critics highlight the poor multi-core scaling in Geekbench 6, which has led to skepticism about its reliability for high-core-count CPUs.
- There's a strong desire for benchmarks that specifically test AI capabilities, indicating that Geekbench doesn't meet the needs of those focused on AI performance metrics.
