LFM2.5 1.2B Instruct on Pixel 8 · 8 GB

Experiment Complete Published

ExecuTorch 1.6.0 on Pixel 8 · 8 GB, CPU

LFM2.5 1.2B Instruct as 8da4w (4-bit), archive 256 input / 128 output: a 256-token prompt and 128 tokens out. Measured 13 Nov 2026, 06:29 UTC by Glenn Sonna, Xybrid.

Eligible for ranking when the objective's required metrics are measured and this configuration is selected.Approximate comparison: The experiments use different effective KV cache precision: fp32, litert-lm default, f16. The experiments here use different 4-bit schemes: 8da4w, litert-int4, Q4_K_M.

Experiment id s4-pixel8-lfm2.5-1.2b-4-bit-et-xnnpack-256

Summary

Over 6 valid trials of 6 attempted (6 planned), computed from the raw trials below.

MetricMedianRangeMean ± SDQuartiles
Decode speed tok/s, higher is better31.7 29.4–32.3 31.4 ± 1.1 31.3–32.0
Time to first token ms, lower is betterNot measured
Peak memory GiB, lower is betterNot measured
Prefill speed tok/s, higher is better156.3 147.2–165.4 156.4 ± 7.3 151.4–161.7

Trials, metric by metric

The line runs from the lowest to the highest valid trial, the dot is the median and each tick is one trial.

Decode speed tok/s, higher is better
Time to first token ms, lower is better
Peak memory GiB, lower is better
Prefill speed tok/s, higher is better

Every trial

Raw token counts and durations, with the speeds computed from them. Invalid trials stay in the record with their reason; they are left out of the summary, never silently dropped.

Every attempted trial
TrialTime to first token msPrefill tok/sDecode tok/sPeak memory GiBThermalNote
1not recorded147.2256 tok in 1,739 ms, engine29.4127 tok in 4,324 ms, enginenot recorded38.3 °C, nominal → 38.3 °C, nominal
2not recorded161.0256 tok in 1,590 ms, engine31.3127 tok in 4,058 ms, enginenot recorded35.4 °C, nominal → 36.3 °C, nominal
3not recorded165.4256 tok in 1,548 ms, engine32.0127 tok in 3,970 ms, enginenot recorded34.1 °C, nominal → 35.2 °C, nominal
4not recorded151.6256 tok in 1,689 ms, engine31.4127 tok in 4,039 ms, enginenot recorded33.9 °C, nominal → 35.8 °C, nominal
5not recorded151.4256 tok in 1,691 ms, engine32.0127 tok in 3,971 ms, enginenot recorded33.5 °C, nominal → 35.5 °C, nominal
6not recorded161.9256 tok in 1,581 ms, engine32.3127 tok in 3,931 ms, enginenot recorded33.5 °C, nominal → 35.5 °C, nominal

Caveats

  • Publisher cards identify the base model but do not always pin the exact export source checkpoint revision.
  • Engines may differ in quantization scheme, effective KV cache precision, kernels, and activations.

Configuration

Device

Device
Google Pixel 8
Cohort
Android phones (mobile)
SoC
Google Tensor G3
CPU
9-core CPU (1× Cortex-X3, 4× Cortex-A715, 4× Cortex-A510)
GPU
Arm Mali-G715
NPU
Not recorded
RAM
8.00 GiB
OS
Android 16 (CP1A.260505.005)
Kernel
6.1.145-android14-11-gfa1d6308d1fe-ab14691759
Power
mains
Low power mode
No
Wakefulness
awake

Engine

Engine
ExecuTorch
Maintainer
PyTorch (Meta)
License
BSD-3-Clause, open source
Version
1.6.0
Commit
4776f71b6f0daa5dfc7d758d5836fc1cc0d40154
Binary SHA-256
a214c779c6d098e58572ee7c3c22494afc81a5ad50cbc0fd68d1049f3d872b5a
Built from
source
Build flags
None
Backend
CPU
Threads
Not recorded
Interface
cli
Settings in effect
backend=cpu recipe=lfm2.5-1.2b-et-xnnpack-8da4w,lfm2_xnnpack_q8da4w.yaml,lfm2.5-1.2b-et-xnnpack-8da4w.pte runner=xybrid_et_runner process=one per generation echo=false temperature=0 num_bos=0 num_eos=0 reset_before_generation=true eos_ignorable=false threads=engine default (performance cores)

Model and weights

Base model
LFM2.5 1.2B Instruct (LFM2.5, 1.2B)
Upstream
LiquidAI/LFM2.5-1.2B-Instruct@b9d816ab510f9a28dca37cb182e9da8b0226e13c
Weights
LiquidAI/LFM2.5-1.2B-Instruct@b9d816ab510f9a28dca37cb182e9da8b0226e13c
Format
pte
File 1
lfm2.5-1.2b-et-xnnpack-8da4w.pte · 1.07 GiB · sha256 4052582dd48e85b9980273d2092c0009fb4bd67356f7011d1e2bf2757d26a9cd
File 2
tokenizer.json · 4.5 MiB · sha256 df1d8d5ec5d091b460562ffd545e4a5e91d17d4a0db7ebe733be34ed374377bd
Quantization
8da4w (4-bit)
Weight bits
4
Group size
Not recorded
KV cache
fp32
Conversion
Converted by Xybrid with executorch export_llm 1.6.0
Conversion command
cd "$ROOT/engines/executorch/build/lfm2.5-1.2b-et-xnnpack-8da4w" && PYTHONPATH="$ROOT/engines/executorch/.venv/lib/python3.12/site-packages/executorch" "$ROOT/engines/executorch/.venv/bin/python" -m executorch.extension.llm.export.export_llm --config "$ROOT/engines/executorch/.venv/lib/python3.12/site-packages/executorch/examples/models/lfm2/config/lfm2_xnnpack_q8da4w.yaml" +base.model_class=lfm2_5_1_2b "+base.checkpoint=$ROOT/engines/executorch/build/lfm2.5-1.2b-weights.pth" "+base.params=$ROOT/downloads/executorch/4776f71b6f0daa5dfc7d758d5836fc1cc0d40154/examples/models/lfm2/config/lfm2_5_1_2b_config.json" +export.max_seq_length=4096 +export.max_context_length=4096 +export.output_name=lfm2.5-1.2b-et-xnnpack-8da4w.pte

Workload

Workload
Archive 256 input / 128 output
Prompt
256 tokens, from prompts/source.txt
Prompt SHA-256
c10d1b7bef62d95322f8f8d229dc66118d9375107bdc49b06f76287d058b2f6c
Chat template
ba551d58630afa3190b1be3602e28301f3d2e9bbac978dfc49d6d825171648b6
Output
128 tokens, ignoring end of sequence
Context length
4,096
Batch size
1
Sampling
Greedy
Timing
One streamed request per trial
Warmup
2 runs, not counted
Trials planned
6
Cooldown
60 s between trials

Conditions

Ambient
Not recorded
Thermal at start
34.3 °C, nominal
Thermal at end
35.5 °C, nominal
Thermal gate
Android thermal status none (0)
Throttling detected
No

Provenance

Measured by
Glenn Sonna, Xybrid
Operator affiliation
Xybrid
Run at
13 Nov 2026, 06:29 UTC
Harness
xybrid-bench 0.4.0 (0df92fb167d0c33b8cb41e526be016e53912b7b2)
Methodology
v0.6
Publication
published
Published
16 Nov 2026, 10:23 UTC
Reviewed by
Glenn Sonna

Output check

Result
Passed
Method
sanity-v1-bf16-reference
Reference
checks/sanity-v1.json
Reference SHA-256
a702dee079525ae0ec90465b1c9e26052c99cb3e1c53e83dea05be4c194fc7ca
Required passing checks
11
Check arithmetic
Passed · sha256 6b51d431df5d7f141cbececcf79edf3dd861c3b4069f0b11661a3eefacbba918
Check capital
Passed · sha256 5dd272b4f316b776a7b8e3d0894b37e1e42be3d5d3b204b8a5836cc50597a6b1
Check opposite
Passed · sha256 f151cd9e12c8994edfa316e4624f956bc68460454869177f287146f8cc956a12
Check weekday
Passed · sha256 7d8af1de1262f150187d5938bf649d6c35726970685f916d1ec6392a801b9762
Check color
Failed · sha256 0d5adda3cdccf937bd9c9e66c8000f98d9d0a91e7c3a166e2dfdc91386911cd2
Check sort
Passed · sha256 a386b3c9b4b4786df5bb6474bab0a62b8476e2f3f9c8a6433aca40152840f6b7
Check animal
Passed · sha256 4b227777d4dd1fc61c6f884f48641d02b4d121d3fd328cb08b5531fcacdabf8a
Check word
Passed · sha256 3b635d3b4437e023e72a9df26201b6ab51285b376446446ebb379121994550f6
Check copy
Passed · sha256 d61e6890d0e3571b0c7693612d41f8d059a1fcdef4ba959c0af4144541d41907
Check needle-start
Passed · sha256 8b9af2e5b8b4cfddb87d171997b95c471f8d68a68e30fb5177abeb3332372edd
Check needle-end
Passed · sha256 e6589c736af32f81bf239c5eda6de41959122ccc1d05463b75e0eb0c70fc831c
Check continuation-128
Passed · sha256 fe1fdf5d49caa68c1ef2f8a341413cb2c3826023aef1113c6c289dcd86193b32
Detail
11/12 of the checks the model passes at BF16; needs 11 and the mandatory ones (needle-start, needle-end, continuation-128). Left out, failed by every BF16 configuration: none.

Reproduce it

Commands
pnpm bench run --plan plans/s4-pixel8-lfm2.5-1.2b-4-bit.json

Three palindromic blocks per quantization class and workload; load, two warmups, one trial, unload per turn.

Raw data