Experiment Complete Published
Ollama 0.40.13 on M5 Max · 64 GiB, Metal
Ternary Bonsai 2 27B as Q2_0 (2-bit), archive 2048 input / 128 output: a 2048-token prompt and 128 tokens out. Measured 9 Nov 2026, 23:19 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-ollama-2048
Summary
Over 6 valid trials of 6 attempted (6 planned), computed from the raw trials below.
| Metric | Median | Range | Mean ± SD | Quartiles |
|---|---|---|---|---|
| Decode speed tok/s, higher is better | 44.6 | 44.2–44.6 | 44.5 ± 0.2 | 44.5–44.6 |
| Time to first token ms, lower is better | 2,672 | 2,572–2,941 | 2,703 ± 124 | 2,668–2,689 |
| Peak memory GiB, lower is better | 8.12 | 8.12–8.12 | 8.12 ± 0.00 | 8.12–8.12 |
| Prefill speed tok/s, higher is better | 767.8 | 697.4–798.1 | 760.3 ± 33.4 | 762.9–768.8 |
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.
| Trial | Time to first token ms | Prefill tok/s | Decode tok/s | Peak memory GiB | Thermal | Note |
|---|---|---|---|---|---|---|
| 1 | 2,572 | 798.12,048 tok in 2,566 ms, engine | 44.6127 tok in 2,849 ms, engine | 8.12 | nominal → nominal | |
| 2 | 2,668 | 768.92,048 tok in 2,663 ms, engine | 44.5127 tok in 2,852 ms, engine | 8.12 | nominal → nominal | |
| 3 | 2,674 | 767.12,048 tok in 2,670 ms, engine | 44.6127 tok in 2,846 ms, engine | 8.12 | nominal → nominal | |
| 4 | 2,694 | 761.52,048 tok in 2,689 ms, engine | 44.6127 tok in 2,848 ms, engine | 8.12 | nominal → nominal | |
| 5 | 2,941 | 697.42,048 tok in 2,937 ms, engine | 44.2127 tok in 2,872 ms, engine | 8.12 | nominal → nominal | |
| 6 | 2,670 | 768.52,048 tok in 2,665 ms, engine | 44.6127 tok in 2,845 ms, engine | 8.12 | nominal → 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
- Ollama
- Maintainer
- Ollama
- License
- MIT, open source
- Version
- 0.40.13
- Commit
- Not recorded
- Binary SHA-256
- 270089e59e1d411ac4d0987154b4e95f48b29a81c9d274833c6b766d6df88300
- Built from
- release
- Build flags
- None
- Backend
- Metal
- Threads
- Not recorded
- Interface
- http-server
- Settings in effect
- backend=metal model=xybrid-4f99aed01b8a877e blob=sha256-4f99aed01b8a877e153f9aa6569a4440fe17c59701cc0953e36d7f460549e70e offloaded_layers=65 bos_stripped=false llama_server.host=127.0.0.1 llama_server.no-webui=true llama_server.offline=true llama_server.c=4096 llama_server.np=1 llama_server.log-verbosity=4 llama_server.no-log-prefix=true llama_server.no-log-timestamps=true llama_server.cache-type-k=f16 llama_server.cache-type-v=f16 llama_server.flash-attn=on llama_server.b=512 llama_server.ub=512 llama_server.context-shift=true llama_server.keep=4 env.OLLAMA_KEEP_ALIVE=-1 env.OLLAMA_FLASH_ATTENTION=1 env.OLLAMA_KV_CACHE_TYPE=f16 env.OLLAMA_NOHISTORY=1 env.OLLAMA_NUM_PARALLEL=1 env.OLLAMA_MAX_LOADED_MODELS=1 env.LLAMA_ARG_CACHE_RAM=0
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 2048 input / 128 output
- Prompt
- 2,048 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
- fair
- 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, 23:19 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
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
- This record, as committed JSON, schema v0.3.0
- Complete finalized session evidence raw ·
sha256 41452a1391ce…