Version 0.2.0 synchronizes applicable prompt and result fixes through upstream 4aa6761 (source version 0.3.23). Both Core ML and ANE runtimes accept custom noul display labels ({"false": "no", "true": "yes"}), validate questions with question-specific errors, and preserve Unicode structured instructions. Long conversation lists retain their newest tokens; strings and objects retain their beginning. usage reports state truncation and, when present, collapsed options. answer_confidence is the highest answer probability; existing confidence semantics remain unchanged. Neither field establishes calibration accuracy. The checkpoint temperature clamp documented below is included in this release. Exported graph capacity limits still raise errors rather than silently clipping.

Maintenance follows upstream Laya fixes applicable to these runtimes. New runtime features and backend-specific optimization proposals require an upstream-aligned implementation and validation; closing an issue does not establish that its reported behavior is fixed.

Open-weight typed decisions on Apple Silicon. Core ML, Neural Engine, zero generated tokens.

PyPI · Hugging Face weights · 中文

A real Laya model plays Snake locally, with visible probabilities, score, length, latency and safety interventions. The GIF replays a recorded Core ML run at 1× speed. The game uses explicit planner features and a visible cycle safety layer.

The complete active Snake loop sustained 49.1–50.0 decisions/s across three uncapped 600-step episodes, with zero deaths and two safety interventions. Game-loop timings and paced-rate limits include rendering serialization; terminal painting is excluded.

One short multilingual decision: 4.98 ms P50 / 5.31 ms P95 on M3 Max with ANE FP16. The same experiment measured 2.78× better whole-system energy per decision than compiled MLX FP16. A separately validated W8 palette variant reached 4.88 ms and 3.19× energy improvement. These are single-question results, not full Snake frame times; the requested 10× improvement was not achieved.

Apple Silicon · macOS 15+ · Python 3.11–3.13.

pip install 'laya-coreml[demo]'
hf download aac6fef/laya-multilingual-coreml-ane --local-dir models/snake
laya-coreml-snake --model ./models/snake

Download once, then play offline. No PyTorch, Transformers or MLX is needed for inference. The terminal needs 104 columns × 35 rows. Space pauses; ↑/↓ changes speed; R resets; Q quits. First-time Core ML initialization can take tens of seconds.

Controls, recording and video export · Measured stable decision rates · Shareable video and recording provenance

pip install laya-coreml

import laya_coreml as laya

agent = laya.load("aac6fef/laya-multilingual-coreml-ane")
result = agent.predict(
    "The customer requests a refund of a duplicate payment.",
    {
        "refund": {
            "type": "noul",
            "instructions": "Does the customer request a refund?",
        }
    },
)
print(result["answers"]["refund"])

Laya returns probabilities for choice, ordinal score, and boolean noul questions. There is no autoregressive decoding or generated JSON to parse. Hub models download before initialization; subsequent predictions stay local. Pass local_files_only=True to require an existing cache, or load a local directory.

Following upstream v0.3.5, fitted calibration temperatures are clamped to [0.5, 5.0] before use: the shipped choice:11+ bucket is 0.1006, which would sharpen logits ~10x and report a coin flip as near-certainty. The checkpoint's raw values remain available as agent.temperature_raw and agent.temperature_by_options_raw, and a RuntimeWarning names every clamped bucket at load.

The ANE bundle has a 96-token total limit, including question, options and state. Longer requests raise a capacity error. Use aac6fef/laya-multilingual-coreml for the general-purpose 1024-token model. Full API, model selection and offline usage.

40-core GPU, 128 GiB, macOS 27.2. One 91-token question padded to 96, including prompt preparation, tokenization, arrays, synchronous inference, calibration and formatting. Loading and warmup are excluded. MLX enables compile, prefix caching and shape buckets. Six alternating 20-second blocks per implementation produced 65,598 stable calls.

Energy uses direct SMC PSTR sensor readings, with raw samples and explicit anomaly rejection. This is an estimate with sensor and background-load uncertainty. Speed gain × average power ratio = energy gain; multiplying energy by speed again would double-count time. The W8 variant compresses weights while retaining FP16 compute. It is approximate, and its package-size reduction is not a speed ratio.

Speed, energy and hardware evidence · Raw measurements.

Every bundle includes tokenizer/configuration, model card, provenance, checksums and packaging-time validation. ANE bundles also include the exact original host embedding/action tensors they need. No original training checkout is required.

The three general-purpose FP16 checkpoints match upstream selected answers on 189/189 validation questions. Each passes 100 repeated calls. ANE FP16 L96 passes 59/59 fitting questions, with maximum calibrated-probability drift 0.002925; W8 passes the same subset with drift 0.014393 under an unchanged 0.02 gate. Six- and four-bit experiments failed that gate and are not published weights. These are conversion-fidelity fixtures, not proof of general task accuracy.

A separately exported FP16 ANE L1024 graph passes the complete 63/63 fixture, but an actual 1024-token request takes about 91.7 ms in its serial screen. The short ANE result does not establish a long-context advantage. A 600-step paired Snake check matches 600/600 actions, with zero deaths and zero shield interventions; the current ANE adapter's three sequential calls do not establish a consistent full-game speedup over compiled MLX.

The ordinary SDPA Core ML export and the ANE graph are different implementations. The ordinary export defaults to CPU+GPU after unrestricted RangeDim GPU shapes failed local fidelity checks. Changing its device setting alone does not reproduce the ANE result. The ANE rewrite uses BC1L activations, 1×1 projections and per-head attention; its plan and a separate Instruments trace support Neural Engine work. CPU still handles input/output boundaries.

• Install and Python/CLI API
• Snake demo and media
• Release artifacts and pinned Hub revisions
• General Core ML benchmarks
• ANE engineering experiments
• Mathematical investigation of the 10× target
• Conversion issues and supported shapes

To export yourself, install laya-coreml[convert] and run laya-coreml convert laya-multilingual models/custom. ANE research conversion, compression and benchmark scripts live in the Git checkout. The inference wheel contains the portable runtime and optional terminal demo.

Apache-2.0. Independent port of Laya, by Convai Innovations and contributors, building on the MLX sibling project. Not an official Convai Innovations or Apple release. See NOTICE.