Ternary Bonsai 2 27B on M5 Max · 64 GiB

Experiment Complete Published

llama.cpp b12110 on M5 Max · 64 GiB, Metal

Ternary Bonsai 2 27B as Q2_0 (2-bit), archive 256 input / 128 output: a 256-token prompt and 128 tokens out. Measured 9 Nov 2026, 23:06 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: f16, model-dtype.

Experiment id s2-bonsai2-27b-2-bit-llama-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 better43.2 40.8–43.5 42.9 ± 1.1 43.1–43.5
Time to first token ms, lower is better384 382–407 388 ± 9.4 383–388
Peak memory GiB, lower is better7.94 7.94–8.12 7.97 ± 0.07 7.94–7.94
Prefill speed tok/s, higher is better668.1 630.7–670.8 661.2 ± 15.5 661.5–669.7

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.

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
1407630.7256 tok in 406 ms, engine40.8127 tok in 3,117 ms, engine8.12nominal → nominal
2382670.8256 tok in 382 ms, engine43.5127 tok in 2,917 ms, engine7.94nominal → nominal
3383669.8256 tok in 382 ms, engine43.1127 tok in 2,943 ms, engine7.94nominal → nominal
4385666.9256 tok in 384 ms, engine43.1127 tok in 2,946 ms, engine7.94nominal → nominal
5389659.7256 tok in 388 ms, engine43.3127 tok in 2,933 ms, engine7.94nominal → nominal
6383669.3256 tok in 382 ms, engine43.5127 tok in 2,917 ms, engine7.94nominal → 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
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
llama.cpp
Maintainer
ggml-org
License
MIT, open source
Version
b12110
Commit
cd959bae9
Binary SHA-256
e7e3c60a5e612460d095bfd219047040d6bb9d0b56466aae30a480980b676959
Built from
release
Build flags
None
Backend
Metal
Threads
18
Interface
http-server
Settings in effect
context_length=4096 parallel=1 speculative=false backend=metal launch.log_verbosity=4 launch.log_colors=off launch.cache_prompt=false launch.cache_ram_mib=0 launch.cache_reuse=0 launch.continuous_batching=false launch.context_shift=false launch.flash_attn=on launch.cache_type_k=f16 launch.cache_type_v=f16 launch.gpu_layers=999 launch.load_mode=mmap offloaded_layers=65 launch.mmap=true n_batch=2048 n_ubatch=512 n_threads=18 n_threads_batch=18 flash_attn=enabled

Model and weights

Base model
Ternary Bonsai 2 27B (Bonsai, 27B)
Upstream
prism-ml/Ternary-Bonsai-2-27B-gguf@839bd57023c1558945afea70a74fa83c054f35f9
Weights
prism-ml/Ternary-Bonsai-2-27B-gguf@839bd57023c1558945afea70a74fa83c054f35f9
Format
gguf
File 1
README.md · 24,367 bytes · sha256 57fa037cfa812c9e6872e970c963d5227904eade7dda064019ae2a3fd140b500
File 2
Ternary-Bonsai-2-27B-Q2_0.gguf · 7.10 GiB · sha256 4f99aed01b8a877e153f9aa6569a4440fe17c59701cc0953e36d7f460549e70e
Quantization
Q2_0 (2-bit)
Weight bits
2
Group size
Not recorded
KV cache
f16
Conversion
Published upstream by prism-ml
Conversion command
Not recorded

Workload

Workload
Archive 256 input / 128 output
Prompt
256 tokens, from prompts/source.txt
Prompt SHA-256
c10d1b7bef62d95322f8f8d229dc66118d9375107bdc49b06f76287d058b2f6c
Chat template
c3cf9e34abf4f9e36c2d72165aa9c132d3e2a725b6c2586aaa3a8af9d7a81041
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
No

Provenance

Measured by
Glenn Sonna, Xybrid
Operator affiliation
Xybrid
Run at
9 Nov 2026, 23:06 UTC
Harness
xybrid-bench 0.4.0 (103beb8cf54e7e006744c19f3189370cd245047e)
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 e272deecde720ffc773d7a6a81adabbc366ccbb64da5abdb6b26f52cd3f952cb
Check weekday
Passed · sha256 7d8af1de1262f150187d5938bf649d6c35726970685f916d1ec6392a801b9762
Check color
Passed · sha256 d486dfbd5fb578340bccbdd0a662527eab38550648d5f44517e5ac71b8824703
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 ed26731fe8d39dacac77703998aa4bb784317ae565cb21a58f83f75d7e458142
Detail
12/12; needs 11 and every mandatory check (needle-start, needle-end, continuation-128); the model has no BF16 reference.

Reproduce it

Commands
pnpm bench run --plan plans/s2-bonsai2-27b-2-bit.json

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

Raw data