Experiment Complete Published
MLX LM 0.32.2 on M5 Max · 64 GiB, Metal
LFM2.5 1.2B Instruct as BF16 (bf16), archive 2048 input / 128 output: a 2048-token prompt and 128 tokens out. Measured 12 Nov 2026, 19:57 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, magnitude default, model-dtype, zml default.
Experiment id s3-lfm2.5-1.2b-bf16-mlx-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 | 198.3 | 197.8–199.2 | 198.4 ± 0.5 | 198.1–198.8 |
| Time to first token ms, lower is better | 94.2 | 93.9–95.0 | 94.3 ± 0.5 | 94.0–94.6 |
| Peak memory GiB, lower is better | 3.26 | 3.25–3.26 | 3.25 ± 0.00 | 3.25–3.26 |
| Prefill speed tok/s, higher is better | 21,872 | 21,689–21,956 | 21,844 ± 105.6 | 21,772–21,919 |
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 | 93.9 | 21,9562,048 tok in 93.3 ms, engine | 198.4127 tok in 640 ms, engine | 3.26 | nominal → nominal | |
| 2 | 94.7 | 21,7472,048 tok in 94.2 ms, engine | 197.8127 tok in 642 ms, engine | 3.26 | nominal → nominal | |
| 3 | 94.3 | 21,8472,048 tok in 93.7 ms, engine | 199.2127 tok in 638 ms, engine | 3.26 | nominal → nominal | |
| 4 | 94.0 | 21,9272,048 tok in 93.4 ms, engine | 198.9127 tok in 639 ms, engine | 3.26 | nominal → nominal | |
| 5 | 94.1 | 21,8962,048 tok in 93.5 ms, engine | 198.3127 tok in 640 ms, engine | 3.25 | nominal → nominal | |
| 6 | 95.0 | 21,6892,048 tok in 94.4 ms, engine | 198.0127 tok in 641 ms, engine | 3.25 | 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
- MLX LM
- Maintainer
- Apple (ml-explore)
- License
- MIT, open source
- Version
- 0.32.2
- Commit
- Not recorded
- Binary SHA-256
- 9d46803799f99d6fc7c35fff2d4357c79b1a25ae98cc63efeaaebeb50a7b5f33
- Built from
- package
- Build flags
- None
- Backend
- Metal
- Threads
- Not recorded
- Interface
- python-worker
- Settings in effect
- backend=Device(gpu, 0) prompt_cache=false speculative=false greedy=true temperature=0 sampler=argmax prefill_step_size=2048 context_length=4096 kv_quantized=false batch_size=1 trust_remote_code=false
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-MLX-bf16@fed5ca98f2edd9817149d4becfa239d70c2fde4b
- Format
- safetensors
- File 1
- README.md · 2,203 bytes · sha256 76f229cf9d76aa3bd80323b31d61abb9e9a043b262ed2758a609e8aae8cf32a0
- File 2
- chat_template.jinja · 1,836 bytes · sha256 3acb41f339d069a43037e6b9a9715cbeece8874eb25a7cabf1a33903d7724d6d
- File 3
- config.json · 1,367 bytes · sha256 1423345cee34bec410b5463c415b0748141d8360bbe33c7bce9e93da2526c65c
- File 4
- generation_config.json · 132 bytes · sha256 5ffd97da1dec4308543894569662d96e923ed01f7a9d8c7ff5aea7f800738cbd
- File 5
- model.safetensors · 2.18 GiB · sha256 a976ae11300d40c9796fe8b3e5276f17fd4e28a3acbc55c62cb6c07d6c670f71
- File 6
- model.safetensors.index.json · 10,375 bytes · sha256 b541548d92fbb369053287da0f13a567de2e6685c121c138bfd6d84673db92d2
- File 7
- special_tokens_map.json · 434 bytes · sha256 742aefe2b7dec496e8caffdba03a75d0c1a9925d53bd3f3e0d388c96b591b6f4
- File 8
- tokenizer.json · 4.5 MiB · sha256 df1d8d5ec5d091b460562ffd545e4a5e91d17d4a0db7ebe733be34ed374377bd
- File 9
- tokenizer_config.json · 92,225 bytes · sha256 2a52ec012d3df831ba434b081bef3726a6ee22501f062ad8353c557a0cfa0d01
- Quantization
- BF16 (bf16)
- Weight bits
- 16
- Group size
- Not recorded
- KV cache
- model-dtype
- 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
- 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
- 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
- 12 Nov 2026, 19:57 UTC
- Harness
- xybrid-bench 0.4.0 (125f448ac6369993822f981281606bb885381311)
- 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
- 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 fe1fdf5d49caa68c1ef2f8a341413cb2c3826023aef1113c6c289dcd86193b32
- Detail
- 12/12 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
pnpm bench run --plan plans/s3-lfm2.5-1.2b-bf16.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 829c794f9e2c…