Basic information
- Board URL (official): https://frame.work/products/laptop12-diy-intel-13gen
- Board purchased from: Framework
- Board purchase date: 2026-05-17
- Board specs (as tested): DIY Edition, w/ Power Adapter, i3-1315U, added 8GB RAM, 256GB NVMe SSD
- Board price (as tested): $749
Linux/system information
# output of `screenfetch`
./+o+- jgeerling@framework-12
yyyyy- -yyyyyy+ OS: Ubuntu 26.04 resolute
://+//////-yyyyyyo Kernel: x86_64 Linux 7.0.0-15-generic
.++ .:/++++++/-.+sss/` Uptime: 2m
.:++o: /++++++++/:--:/- Packages: 1707
o:+o+:++.`..```.-/oo+++++/ Shell: dash
.:+o:+o/. `+sssoo+/ Disk: 12G / 234G (6%)
.++/+:+oo+o:` /sssooo. CPU: 13th Gen Intel Core i3-1315U @ 8x 4.5GHz [55.0°C]
/+++//+:`oo+o /::--:. GPU: UHD Graphics
\+/+o+++`o++o ++////. RAM: 1666MiB / 6892MiB
.++.o+++oo+:` /dddhhh.
.+.o+oo:. `oddhhhh+
\+.++o+o``-````.:ohdhhhhh+
.o:`.syhhhhhhh/.oo++o`
`oo++.
.+.o+oo:. `oddhhhh+
\+/+o+++`o++o ++////.
.++ .:/++++++/-.+sss/`
# output of `uname -a`
Linux framework-12 7.0.0-15-generic #15-Ubuntu SMP PREEMPT_DYNAMIC Wed Apr 22 16:06:43 UTC 2026 x86_64 GNU/Linux
System topology
Benchmark results
CPU
Power
- Idle power draw (at wall): 4.2 W (11.5W full brightness, 1.1W running with screen off)
- Maximum simulated power draw (
stress-ng --matrix 0): 23W (with early spike to 28.7W)
- During Geekbench multicore benchmark: 54 W
- During
top500 HPL benchmark: 24 W (early spike to 46.3W, but throttles quickly)
Note: All power benchmarks taken from the wall using Framework's official 60W GaN charger after charging the battery to 100%, with screen at 50% brightness, in the Performance profile in Ubuntu 26.04, waiting at least 1 hour for charge rates to settle to < 0.3W.
Disk
KBG40ZNS256G NVMe KIOXIA 256GB
| Benchmark |
Result |
| iozone 4K random read |
69.40 MB/s |
| iozone 4K random write |
338.39 MB/s |
| iozone 1M random read |
1827.95 MB/s |
| iozone 1M random write |
1520.70 MB/s |
| iozone 1M sequential read |
1830.12 MB/s |
| iozone 1M sequential write |
1506.14 MB/s |
Network
iperf3 results:
iperf3 -c $SERVER_IP: 1.36 Gbpsiperf3 -c $SERVER_IP --reverse: 903 Mbpsiperf3 -c $SERVER_IP --bidir: 302 Mbps up, 1.21 Gbps down
Tested with built-in AX201 WiFi (Intel Raptor Lake PCH CNVi WiFi), on 802.11be WiFi 7 6 GHz network, with 6' distance to AP.
GPU
glmark2
glmark2-es2 / glmark2-es2-wayland results:
=======================================================
glmark2 2023.01
=======================================================
OpenGL Information
GL_VENDOR: Intel
GL_RENDERER: Mesa Intel(R) Graphics (RPL-U)
GL_VERSION: OpenGL ES 3.2 Mesa 26.0.3-1ubuntu1
Surface Config: buf=32 r=8 g=8 b=8 a=8 depth=24 stencil=0 samples=0
Surface Size: 800x600 windowed
=======================================================
[build] use-vbo=false: FPS: 3957 FrameTime: 0.253 ms
[build] use-vbo=true: FPS: 4091 FrameTime: 0.244 ms
[texture] texture-filter=nearest: FPS: 4288 FrameTime: 0.233 ms
[texture] texture-filter=linear: FPS: 4310 FrameTime: 0.232 ms
[texture] texture-filter=mipmap: FPS: 4232 FrameTime: 0.236 ms
[shading] shading=gouraud: FPS: 5732 FrameTime: 0.174 ms
[shading] shading=blinn-phong-inf: FPS: 5599 FrameTime: 0.179 ms
[shading] shading=phong: FPS: 5958 FrameTime: 0.168 ms
[shading] shading=cel: FPS: 5770 FrameTime: 0.173 ms
[bump] bump-render=high-poly: FPS: 5104 FrameTime: 0.196 ms
[bump] bump-render=normals: FPS: 10378 FrameTime: 0.096 ms
[bump] bump-render=height: FPS: 7523 FrameTime: 0.133 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 6818 FrameTime: 0.147 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 3847 FrameTime: 0.260 ms
[pulsar] light=false:quads=5:texture=false: FPS: 8055 FrameTime: 0.124 ms
[desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 3323 FrameTime: 0.301 ms
[desktop] effect=shadow:windows=4: FPS: 5404 FrameTime: 0.185 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 1094 FrameTime: 0.915 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 2402 FrameTime: 0.416 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 1537 FrameTime: 0.651 ms
[ideas] speed=duration: FPS: 5564 FrameTime: 0.180 ms
[jellyfish] <default>: FPS: 6057 FrameTime: 0.165 ms
[terrain] <default>: FPS: 482 FrameTime: 2.079 ms
[shadow] <default>: FPS: 6987 FrameTime: 0.143 ms
[refract] <default>: FPS: 1174 FrameTime: 0.852 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 9477 FrameTime: 0.106 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 9197 FrameTime: 0.109 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 9139 FrameTime: 0.109 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 9101 FrameTime: 0.110 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 9039 FrameTime: 0.111 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 8927 FrameTime: 0.112 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 8763 FrameTime: 0.114 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 8461 FrameTime: 0.118 ms
=======================================================
glmark2 Score: 5810
=======================================================
vkmark
vkmark results:
=======================================================
vkmark 2025.01
=======================================================
Vendor ID: 0x8086
Device ID: 0xA7A9
Device Name: Intel(R) Graphics (RPL-U)
Driver Version: 109051907
Device UUID: 8c4b59c2c09e6c35a87eeb355462de8b
=======================================================
[vertex] device-local=true: FPS: 14959 FrameTime: 0.067 ms
[vertex] device-local=false: FPS: 15243 FrameTime: 0.066 ms
[texture] anisotropy=0: FPS: 14120 FrameTime: 0.071 ms
[texture] anisotropy=16: FPS: 13815 FrameTime: 0.072 ms
[shading] shading=gouraud: FPS: 10875 FrameTime: 0.092 ms
[shading] shading=blinn-phong-inf: FPS: 10575 FrameTime: 0.095 ms
[shading] shading=phong: FPS: 10414 FrameTime: 0.096 ms
[shading] shading=cel: FPS: 10182 FrameTime: 0.098 ms
[effect2d] kernel=edge: FPS: 7980 FrameTime: 0.125 ms
[effect2d] kernel=blur: FPS: 3799 FrameTime: 0.263 ms
[desktop] <default>: FPS: 5506 FrameTime: 0.182 ms
[cube] <default>: FPS: 17972 FrameTime: 0.056 ms
[clear] <default>: FPS: 14999 FrameTime: 0.067 ms
=======================================================
vkmark Score: 11572
=======================================================
Note: vkmark needs to be compiled from source on Debian 12 and earlier.
GravityMark
GravityMark result: 8,501
AI / LLM Inference
Basic AI inference results:
build: 4f0e43da6 (9282)
| model |
size |
params |
backend |
threads |
test |
t/s |
| llama 1B Q4_K - Medium |
636.18 MiB |
1.10 B |
CPU |
2 |
pp512 |
89.69 ± 0.08 |
| llama 1B Q4_K - Medium |
636.18 MiB |
1.10 B |
CPU |
2 |
pp4096 |
55.30 ± 3.60 |
| llama 1B Q4_K - Medium |
636.18 MiB |
1.10 B |
CPU |
2 |
tg128 |
33.14 ± 0.11 |
| llama 1B Q4_K - Medium |
636.18 MiB |
1.10 B |
CPU |
2 |
pp4096+tg128 |
54.59 ± 0.03 |
| model |
size |
params |
backend |
threads |
test |
t/s |
| llama 3B Q4_K - Medium |
1.87 GiB |
3.21 B |
CPU |
2 |
pp512 |
28.67 ± 0.18 |
| llama 3B Q4_K - Medium |
1.87 GiB |
3.21 B |
CPU |
2 |
pp4096 |
23.91 ± 0.04 |
| llama 3B Q4_K - Medium |
1.87 GiB |
3.21 B |
CPU |
2 |
tg128 |
11.05 ± 0.08 |
| llama 3B Q4_K - Medium |
1.87 GiB |
3.21 B |
CPU |
2 |
pp4096+tg128 |
22.55 ± 0.01 |
Memory
tinymembench results:
Click to expand memory benchmark result
tinymembench v0.4.10 (simple benchmark for memory throughput and latency)
==========================================================================
== Memory bandwidth tests ==
== ==
== Note 1: 1MB = 1000000 bytes ==
== Note 2: Results for 'copy' tests show how many bytes can be ==
== copied per second (adding together read and writen ==
== bytes would have provided twice higher numbers) ==
== Note 3: 2-pass copy means that we are using a small temporary buffer ==
== to first fetch data into it, and only then write it to the ==
== destination (source -> L1 cache, L1 cache -> destination) ==
== Note 4: If sample standard deviation exceeds 0.1%, it is shown in ==
== brackets ==
==========================================================================
C copy backwards : 9295.6 MB/s (0.8%)
C copy backwards (32 byte blocks) : 9274.5 MB/s (0.2%)
C copy backwards (64 byte blocks) : 9296.5 MB/s (0.3%)
C copy : 9196.8 MB/s (0.3%)
C copy prefetched (32 bytes step) : 9009.0 MB/s (0.1%)
C copy prefetched (64 bytes step) : 9054.5 MB/s (0.2%)
C 2-pass copy : 8220.8 MB/s (0.6%)
C 2-pass copy prefetched (32 bytes step) : 7630.9 MB/s (0.3%)
C 2-pass copy prefetched (64 bytes step) : 7717.4 MB/s (1.5%)
C fill : 13616.7 MB/s (0.5%)
C fill (shuffle within 16 byte blocks) : 13641.5 MB/s (0.5%)
C fill (shuffle within 32 byte blocks) : 13608.9 MB/s (0.4%)
C fill (shuffle within 64 byte blocks) : 13629.9 MB/s (0.6%)
---
standard memcpy : 13877.8 MB/s (0.5%)
standard memset : 27137.0 MB/s
---
MOVSB copy : 9189.4 MB/s (0.2%)
MOVSD copy : 9195.6 MB/s (4.2%)
SSE2 copy : 9659.5 MB/s (0.2%)
SSE2 nontemporal copy : 13010.0 MB/s (0.4%)
SSE2 copy prefetched (32 bytes step) : 9262.3 MB/s (0.3%)
SSE2 copy prefetched (64 bytes step) : 9270.2 MB/s (0.2%)
SSE2 nontemporal copy prefetched (32 bytes step) : 12446.6 MB/s (0.3%)
SSE2 nontemporal copy prefetched (64 bytes step) : 12493.3 MB/s (0.4%)
SSE2 2-pass copy : 8581.3 MB/s (0.9%)
SSE2 2-pass copy prefetched (32 bytes step) : 8331.3 MB/s (0.4%)
SSE2 2-pass copy prefetched (64 bytes step) : 8295.5 MB/s (1.1%)
SSE2 2-pass nontemporal copy : 4719.7 MB/s (3.7%)
SSE2 fill : 13348.6 MB/s (15.5%)
SSE2 nontemporal fill : 27463.7 MB/s (2.0%)
==========================================================================
== Memory latency test ==
== ==
== Average time is measured for random memory accesses in the buffers ==
== of different sizes. The larger is the buffer, the more significant ==
== are relative contributions of TLB, L1/L2 cache misses and SDRAM ==
== accesses. For extremely large buffer sizes we are expecting to see ==
== page table walk with several requests to SDRAM for almost every ==
== memory access (though 64MiB is not nearly large enough to experience ==
== this effect to its fullest). ==
== ==
== Note 1: All the numbers are representing extra time, which needs to ==
== be added to L1 cache latency. The cycle timings for L1 cache ==
== latency can be usually found in the processor documentation. ==
== Note 2: Dual random read means that we are simultaneously performing ==
== two independent memory accesses at a time. In the case if ==
== the memory subsystem can't handle multiple outstanding ==
== requests, dual random read has the same timings as two ==
== single reads performed one after another. ==
==========================================================================
block size : single random read / dual random read, [MADV_NOHUGEPAGE]
1024 : 0.0 ns / 0.0 ns
2048 : 0.0 ns / 0.0 ns
4096 : 0.0 ns / 0.0 ns
8192 : 0.0 ns / 0.0 ns
16384 : 0.0 ns / 0.0 ns
32768 : 0.0 ns / 0.0 ns
65536 : 0.7 ns / 1.1 ns
131072 : 1.7 ns / 2.2 ns
262144 : 2.2 ns / 2.5 ns
524288 : 3.0 ns / 3.4 ns
1048576 : 4.1 ns / 4.6 ns
2097152 : 8.3 ns / 10.8 ns
4194304 : 12.0 ns / 14.3 ns
8388608 : 18.5 ns / 23.8 ns
16777216 : 54.1 ns / 75.9 ns
33554432 : 85.2 ns / 110.1 ns
67108864 : 102.2 ns / 120.4 ns
block size : single random read / dual random read, [MADV_HUGEPAGE]
1024 : 0.0 ns / 0.0 ns
2048 : 0.0 ns / 0.0 ns
4096 : 0.0 ns / 0.0 ns
8192 : 0.0 ns / 0.0 ns
16384 : 0.0 ns / 0.0 ns
32768 : 0.0 ns / 0.0 ns
65536 : 0.7 ns / 1.1 ns
131072 : 1.7 ns / 2.2 ns
262144 : 2.2 ns / 2.5 ns
524288 : 2.5 ns / 2.6 ns
1048576 : 2.7 ns / 2.7 ns
2097152 : 6.9 ns / 9.3 ns
4194304 : 10.5 ns / 12.8 ns
8388608 : 11.8 ns / 13.9 ns
16777216 : 47.5 ns / 71.5 ns
33554432 : 78.1 ns / 101.2 ns
67108864 : 92.7 ns / 110.3 ns
Core to Core Memory Latency
Phoronix Test Suite
Results from pi-general-benchmark.sh:
- pts/encode-mp3: 5.935 sec
- pts/x264 1080p: 47.73 fps
- pts/x264 4K: 9.97 fps
- pts/phpbench: 1189544
- pts/build-linux-kernel (defconfig): 397.972 sec
Basic information
Linux/system information
System topology
Benchmark results
CPU
Power
stress-ng --matrix 0): 23W (with early spike to 28.7W)top500HPL benchmark: 24 W (early spike to 46.3W, but throttles quickly)Note: All power benchmarks taken from the wall using Framework's official 60W GaN charger after charging the battery to 100%, with screen at 50% brightness, in the Performance profile in Ubuntu 26.04, waiting at least 1 hour for charge rates to settle to < 0.3W.
Disk
KBG40ZNS256G NVMe KIOXIA 256GB
Network
iperf3results:iperf3 -c $SERVER_IP: 1.36 Gbpsiperf3 -c $SERVER_IP --reverse: 903 Mbpsiperf3 -c $SERVER_IP --bidir: 302 Mbps up, 1.21 Gbps downTested with built-in AX201 WiFi (Intel Raptor Lake PCH CNVi WiFi), on 802.11be WiFi 7 6 GHz network, with 6' distance to AP.
GPU
glmark2
glmark2-es2/glmark2-es2-waylandresults:vkmark
vkmarkresults:GravityMark
GravityMark result: 8,501
AI / LLM Inference
Basic AI inference results:
build: 4f0e43da6 (9282)
Memory
tinymembenchresults:Click to expand memory benchmark result
Core to Core Memory Latency
Phoronix Test Suite
Results from pi-general-benchmark.sh: