LFM2.5 2.6B on M5 Max · 64 GiB

Experiment Unsupported Published

ZML 20261106.0 on M5 Max · 64 GiB, Metal

LFM2.5 2.6B as BF16 (bf16), archive 2048 input / 128 output: a 2048-token prompt and 128 tokens out. Measured 10 Nov 2026, 02:09 UTC by Glenn Sonna, Xybrid.

Doesn’t count in rankings: zml can't run the frozen prompts: Error: LLMD streamed reasoning before its answer; the harness can't time it apart from the answer (measured in qualification).

Experiment id s2-lfm2.5-2.6b-bf16-zml-2048

Unsupported configuration

zml can't run the frozen prompts: Error: LLMD streamed reasoning before its answer; the harness can't time it apart from the answer (measured in qualification).

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
Apple Mac17,6
Cohort
Apple silicon Macs (desktop)
SoC
Apple M5 Max
CPU
18-core Apple CPU
GPU
Not recorded
NPU
Not recorded
RAM
64.00 GiB
OS
macOS 27.2.0 (26B123)
Kernel
27.2.0
Power
mains
Low power mode
No
Performance mode
Not recorded

Engine

Engine
ZML
Maintainer
ZML
License
Apache-2.0, open source
Version
20261106.0
Commit
Not recorded
Binary SHA-256
a390b048f1778251624cf4526baf142bad3d60b229cac981dc887a432ca40224
Built from
release
Build flags
None
Backend
Metal
Threads
Not recorded
Interface
http-server
Settings in effect
unsupported=zml can't run the frozen prompts: Error: LLMD streamed reasoning before its answer; the harness can't time it apart from the answer (measured in qualification).

Model and weights

Base model
LFM2.5 2.6B (LFM2.5, 2.6B)
Upstream
LiquidAI/LFM2.5-2.6B@7141b56cd4650af95c7c7b86983443aed050c96a
Weights
LiquidAI/LFM2.5-2.6B@7141b56cd4650af95c7c7b86983443aed050c96a
Format
safetensors
File 1
README.md · 18,300 bytes · sha256 206383f9e7d86fb44fb260474c5a0793830c5691b5ca6f7d5f955058de320886
File 2
chat_template.jinja · 5,443 bytes · sha256 ea663864491de7ade391839479860ca95541f892f72665c73251fbd4643b1bef
File 3
config.json · 1,467 bytes · sha256 480f63fa8e1efa534ae8b92774b3b53b8d6812d62a726e9ecfc866933662f273
File 4
generation_config.json · 327 bytes · sha256 7366b93f26f8830e7a94441a3d2b9f344ceb9e1e31a808e0656df40720075874
File 5
model-00001-of-00002.safetensors · 4.96 GiB · sha256 05be188e6570a7642b4690545ab22ebf4d0a6068775dd776a452eebf738854cc
File 6
model-00002-of-00002.safetensors · 62.0 MiB · sha256 532f803a9e6f127adc124d977503b438d8cc9c18f96e0862bf6835e38867bca8
File 7
model.safetensors.index.json · 21,425 bytes · sha256 1be35d0d99dc56f68e6c5b306f7d4d99b6da97d1ab41ee753832bf1495655c4a
File 8
tokenizer.json · 17.1 MiB · sha256 695be7802a0e4b8a81048f0ff5ebb7fc811a0ba5a6be63dbb24deb5a81096f41
File 9
tokenizer_config.json · 363 bytes · sha256 11f1de897317b489dd09199284528382eeb17f566398b16e2039bb2266d26b09
Quantization
BF16 (bf16)
Weight bits
16
Group size
Not recorded
KV cache
zml default
Conversion
Published upstream by LiquidAI
Conversion command
Not recorded

Workload

Workload
Archive 2048 input / 128 output
Prompt
2,048 tokens, from prompts/source.txt
Prompt SHA-256
c10d1b7bef62d95322f8f8d229dc66118d9375107bdc49b06f76287d058b2f6c
Chat template
ea663864491de7ade391839479860ca95541f892f72665c73251fbd4643b1bef
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
15 s between trials

Conditions

Ambient
Not recorded
Thermal at start
nominal
Thermal at end
nominal
Thermal gate
macOS thermal state nominal
Throttling detected
Not recorded

Provenance

Measured by
Glenn Sonna, Xybrid
Operator affiliation
Xybrid
Run at
10 Nov 2026, 02:09 UTC
Harness
xybrid-bench 0.4.0 (0d9d83cbb53ef082d8c47f706b677721d6241180)
Methodology
v0.6
Publication
published
Published
16 Nov 2026, 10:23 UTC
Reviewed by
Glenn Sonna

Output check

Result
Not run
Method
Not recorded
Reference
Not recorded
Reference SHA-256
Not recorded
Required passing checks
Not recorded
Detail
Not recorded

Reproduce it

Commands
pnpm bench run --plan plans/s2-lfm2.5-2.6b-bf16.json

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

Raw data