Experiment Complete Published
Rapid-MLX 1.0.2 on M5 Max · 64 GiB, Metal
LFM2.5 350M as mlx-affine-4bit-g64 (4-bit), archive 256 input / 128 output: a 256-token prompt and 128 tokens out. Measured 12 Nov 2026, 20:38 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: bf16, fp32, litert-lm default, f16, magnitude default, model-dtype, uzu default. Its weights are 4-bit in another scheme: mlx-affine-4bit-g64, where most use Q4_K_M.
Experiment id s3-lfm2.5-350m-4-bit-rapid-mlx-256
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 | 1,108 | 1,100–1,110 | 1,107 ± 3.8 | 1,106–1,110 |
| Time to first token ms, lower is better | 24.5 | 23.9–25.6 | 24.6 ± 0.6 | 24.4–24.7 |
| Peak memory GiB, lower is better | 0.72 | 0.68–0.74 | 0.72 ± 0.02 | 0.72–0.72 |
| Prefill speed tok/s, higher is better | 10,658 | 10,202–10,949 | 10,620 ± 248.6 | 10,551–10,718 |
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 | 25.6 | 10,202256 tok in 25.1 ms, engine | 1,106127 tok in 115 ms, engine | 0.72 | nominal → nominal | |
| 2 | 24.6 | 10,603256 tok in 24.1 ms, engine | 1,110127 tok in 114 ms, engine | 0.68 | nominal → nominal | |
| 3 | 24.8 | 10,534256 tok in 24.3 ms, engine | 1,100127 tok in 115 ms, engine | 0.72 | nominal → nominal | |
| 4 | 23.9 | 10,949256 tok in 23.4 ms, engine | 1,106127 tok in 115 ms, engine | 0.72 | nominal → nominal | |
| 5 | 24.4 | 10,712256 tok in 23.9 ms, engine | 1,109127 tok in 115 ms, engine | 0.74 | nominal → nominal | |
| 6 | 24.4 | 10,721256 tok in 23.9 ms, engine | 1,110127 tok in 114 ms, engine | 0.72 | 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
- Rapid-MLX
- Maintainer
- Rapid-MLX contributors
- License
- Apache-2.0, open source
- Version
- 1.0.2
- Commit
- f1aa011aa91bb38140ed3fb68f6267b77447ee0f
- Binary SHA-256
- d5c9b638732cda0b2e2a816aa85cc1780e0a3c1b690bb23c8d9b4d3ac81a527c
- Built from
- package
- Build flags
- None
- Backend
- Metal
- Threads
- Not recorded
- Interface
- python-worker
- Settings in effect
- backend=Device(gpu, 0) model_source=artifact folder context_length=4096 batch_size=1 max_num_seqs=1 max_concurrent_requests=1 prefill_batch_size=1 completion_batch_size=1 prefill_step_size=2048 prefix_cache=false disk_caches=false speculative=false spec_decode=none suffix_decoding=false response_cache_entries=0 pflash=off kv_cache_dtype=bf16 kv_quantized=false admission_window_ms=0 stream_interval=1 gpu_memory_utilization=0.9 greedy=true temperature=0 sampler=argmax trust_remote_code=false telemetry=false weight_bits=4 weight_group_size=64
Model and weights
- Base model
- LFM2.5 350M (LFM2.5, 350M)
- Upstream
- LiquidAI/LFM2.5-350M@9921cd4d82c0e3986887aa4b376f9446729ef8a3
- Weights
- LiquidAI/LFM2.5-350M-MLX-4bit@f6cb4e006bb7a2d8a6afa14ec0a53e0586f65a5b
- Format
- safetensors
- File 1
- README.md · 2,329 bytes · sha256 90060ba684dfe9716d391782396bf2084a3941b5bcdb31f1ef6d1b5a5934814e
- File 2
- chat_template.jinja · 2,607 bytes · sha256 1a2d8281b5477abb5bb1cebd69d7d1ee389e074656ca5f5e0ef6ae56d6ae65c1
- File 3
- config.json · 1,714 bytes · sha256 9a8d76fffac255e7aeba7f154fd493e5b42c55b8bf150cc3478484adec6a98ed
- File 4
- generation_config.json · 132 bytes · sha256 1c72d35bb46c5102537eb114fea9a5c10ef928cc813401c3c12aef399c1f8e5d
- File 5
- model.safetensors · 211.5 MiB · sha256 f17b51362f7aa11a8dffafa6f44c625d5d695b2f76122eed11525104810ae809
- File 6
- model.safetensors.index.json · 23,413 bytes · sha256 be372bf2e03a849a8f01821a2e35ffc44840a86d3132ad7e8f2cef5ea655b50f
- File 7
- tokenizer.json · 4.5 MiB · sha256 df1d8d5ec5d091b460562ffd545e4a5e91d17d4a0db7ebe733be34ed374377bd
- File 8
- tokenizer_config.json · 489 bytes · sha256 75c287923e252b08b0a0f1c367bbe557ab23a681d0b71c5a34e0932ddbe2f5ee
- Quantization
- mlx-affine-4bit-g64 (4-bit)
- Weight bits
- 4
- Group size
- 64
- KV cache
- model-dtype
- Conversion
- Published upstream by LiquidAI
- Conversion command
- Not recorded
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
- 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, 20:38 UTC
- Harness
- xybrid-bench 0.4.0 (f67b01e5a1a77cc89f5984f23e35dd4590fa1bec)
- 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
- 7
- Check arithmetic
- Failed · sha256 4a44dc15364204a80fe80e9039455cc1608281820fe2b24f1e5233ade6af1dd5
- Check capital
- Passed · sha256 5dd272b4f316b776a7b8e3d0894b37e1e42be3d5d3b204b8a5836cc50597a6b1
- Check opposite
- Passed · sha256 f151cd9e12c8994edfa316e4624f956bc68460454869177f287146f8cc956a12
- Check weekday
- Passed · sha256 7d8af1de1262f150187d5938bf649d6c35726970685f916d1ec6392a801b9762
- Check color
- Failed · sha256 ba19e9c3d5f49882ddaafed4f286a8a81491150426d321179270e27e74a89097
- Check sort
- Failed · sha256 71f1f3cb483cd6f5ca0ab792084ea71eba3416947c5a687cb3167d6b61af25c7
- Check animal
- Passed · sha256 4b227777d4dd1fc61c6f884f48641d02b4d121d3fd328cb08b5531fcacdabf8a
- Check word
- Failed · sha256 df7e70e5021544f4834bbee64a9e3789febc4be81470df629cad6ddb03320a5c
- Check copy
- Passed · sha256 d61e6890d0e3571b0c7693612d41f8d059a1fcdef4ba959c0af4144541d41907
- Check needle-start
- Passed · sha256 8b9af2e5b8b4cfddb87d171997b95c471f8d68a68e30fb5177abeb3332372edd
- Check needle-end
- Passed · sha256 e6589c736af32f81bf239c5eda6de41959122ccc1d05463b75e0eb0c70fc831c
- Check continuation-128
- Failed · sha256 934273f7da729e4f84dd4eebf5ca05249a63ff43f144a2db708ff023fbb8c7d1
- Detail
- 7/8 of the checks the model passes at BF16; needs 7 and the mandatory ones (needle-start, needle-end). Left out, failed by every BF16 configuration: arithmetic, sort, word, continuation-128.
Reproduce it
pnpm bench run --plan plans/s3-lfm2.5-350m-4-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 53feb733252e…