Framework 12 Platform Tuning For Better Performance Or Power Efficiency

Written by Michael Larabel in Computers on 8 July 2025 at 12:10 PM EDT. Page 4 of 6. 3 Comments.
C-Ray benchmark with settings of Resolution: 1080p, Rays Per Pixel: 16. Performance was the fastest.
C-Ray benchmark with settings of Resolution: 4K, Rays Per Pixel: 16. Performance was the fastest.
7-Zip Compression benchmark with settings of Test: Compression Rating. Performance was the fastest.
7-Zip Compression benchmark with settings of Test: Decompression Rating. Performance was the fastest.

In multi-threaded CPU workloads the performance mode can provide a nice boost to Raptor Lake's capabilities as well but with significant power increases. Or if you care about using a Framework Laptop 12 all day, the power saver mode can help in enhancing overall power efficiency.

Blender benchmark with settings of Blend File: Fishy Cat, Compute: CPU-Only. Performance was the fastest.
Blender benchmark with settings of Blend File: Fishy Cat, Compute: CPU-Only. Performance was the fastest.
Blender benchmark with settings of Blend File: Fishy Cat, Compute: CPU-Only. Performance was the fastest.
OSPRay Studio benchmark with settings of Camera: 3, Resolution: 1080p, Samples Per Pixel: 1, Renderer: Path Tracer, Acceleration: CPU. Performance was the fastest.
OSPRay Studio benchmark with settings of Camera: 3, Resolution: 1080p, Samples Per Pixel: 1, Renderer: Path Tracer, Acceleration: CPU. Performance was the fastest.
OSPRay Studio benchmark with settings of Camera: 3, Resolution: 1080p, Samples Per Pixel: 1, Renderer: Path Tracer, Acceleration: CPU. Performance was the fastest.

The performance mode can make the Framework Laptop 12 more practical for creators with heavy workloads.

FFmpeg benchmark with settings of Encoder: libx265, Scenario: Live. Performance was the fastest.
FFmpeg benchmark with settings of Encoder: libx265, Scenario: Upload. Performance was the fastest.
FFmpeg benchmark with settings of Encoder: libx265, Scenario: Platform. Performance was the fastest.
Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: Llama-3.1-Tulu-3-8B-Q8_0, Test: Text Generation 128. Performance was the fastest.
Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: Llama-3.1-Tulu-3-8B-Q8_0, Test: Text Generation 128. Performance was the fastest.
Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: Llama-3.1-Tulu-3-8B-Q8_0, Test: Text Generation 128. Performance was the fastest.
Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: granite-3.0-3b-a800m-instruct-Q8_0, Test: Text Generation 128. Performance was the fastest.
Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: granite-3.0-3b-a800m-instruct-Q8_0, Test: Text Generation 128. Performance was the fastest.
Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: granite-3.0-3b-a800m-instruct-Q8_0, Test: Text Generation 128. Performance was the fastest.

At the end of the day though you are still constrained by Raptor Lake and the Core i5 1334U simply having two P cores and eight E cores. Creators and other power users with heavy workloads would be better off with the Framework 13 with AMD Strix Point or other laptop options with more meaningful out-of-the-box performance.

Related Articles