visible-cx/LFM2.5-1.2B-CoreAI

LFM2.5-1.2B-Instruct — Core AI (.aimodel)

LiquidAI/LFM2.5-1.2B-Instruct converted to Core AI .aimodel bundles for Apple silicon by visible-cx. These are derivative artifacts: Liquid AI's weights re-expressed as a Core AI graph with int8 block-32 symmetric weight quantization and a two-entrypoint (decode + chunked-prefill) function map. They load through Core AI on macOS and are not usable by PyTorch, GGUF or MLX.

Provenance

Base checkpoint	LiquidAI/LFM2.5-1.2B-Instruct
Recipe	export_lfm2_multifunction.py int8hu --head-sym --chunk 64
Toolchain base	apple/coreai-models @ b1cb71b8522d99408059fa0b98b8742171bcb0b8 + the coreai-model-zoo python overlay
Toolchain	coreai-torch 0.4.1, coreai-core 1.0.0b2, coreai-opt 0.2.1, torch 2.9.0
Producer fingerprint	coreai-core 1.0.0b2 on every inner <name>.aimodel/metadata.json
Weight format	int8, per-K-block-32, symmetric; symmetric head (--head-sym)
Vocab	65,536
Export functions	main (S=1 decode) + prefill (S=64 chunked prefill), function_map: {"main": ["main", "prefill"]}, weights shared

mf64 in the bundle name means multifunction with a 64-wide prefill.

This model is not a thinking model and needs no chat-template adjustment — its template ends at <|im_start|>assistant\n.

Contents

Path	Bytes	Manifest context	Functions
gpu-pipelined/lfm2_5_1_2b_instruct_decode_int8hu_block32_sym_mf64	1,702,039,167	4096	main + prefill
ctx8192/gpu-pipelined/lfm2_5_1_2b_instruct_decode_int8hu_block32_sym_mf64	1,702,039,171	8192	main + prefill
ctx16384/gpu-pipelined/lfm2_5_1_2b_instruct_decode_int8hu_block32_sym_mf64	1,702,039,167	16384	main + prefill

Each folder holds <name>.aimodel/ (main.mlirb ≈ 1.70 GB, main.hash, asset metadata.json), a bundle-level metadata.json, and tokenizer/ (tokenizer.json, tokenizer_config.json, special_tokens_map.json, chat_template.jinja).

The three folders hold the same weights and the same graph — identical function signatures, identical state descriptors, identical export peak RSS at 4096/8192/16384. --max-ctx sets one manifest integer, language.max_context_length, and nothing else. The folders exist so each declared window is a clean, fingerprinted artifact rather than a hand-edited manifest.

Stop token: eos_token = "<|im_end|>" in all three folders. Clean self-stop on every measured sample.

Requirements

• Apple silicon Mac, Core AI runtime.
• Engine contract: 2 inputs (input_ids, position_ids) → logits. No static inputs, no per-step mask. Runs on the pipelined engine and on the sequential engine, which is the only logits-capable one and therefore the only path for grammar-constrained decoding.
• States: keyCache / valueCache Float16, 6 × 1 × 8 × ? × 64 plus convState Float16, 10 × 1 × 2048 × 2. Dynamic sequence dim → GrowingKVCache (initial 256, doubling), not a static allocation at the manifest maximum. convState does not scale with context.
• KV cost: 12,288 bytes per token (fp16) — 50 MB at 4096, 101 MB at 8192, 201 MB at
• Memory envelope, 16 GB Mac (Metal working set ≈ 10.7 GB): weights 1.70 GB resident; KV is nowhere near binding at any context this bundle declares. Measured max RSS 3.42 GB on guided decoding and 3.46 GB at 15k context; peak in-process footprint 0.14 GB and 0.32 GB respectively. Minimum practical machine memory: 8 GB.
• The bundle manifest declares runtime_env COREAI_CHUNK_THRESHOLD=1.

Performance

Measured on a 16 GB Apple silicon Mac (M2 Pro, macOS 27 beta): 10-sample harness, guided JSON-constrained decoding against a fixed schema, greedy, sequential engine, reset() between samples, 128-token cap. Load excluded from s/row; sample 1 excluded as a cache-warm outlier.

Cold load	5.4 s
Guided JSON parse	10/10
Enum-clean	10/10
s/row (long samples)	1.89
s/row (short samples)	1.37
Decode	71.5–74.7 tok/s
TTFT	0.37–1.60 s
Peak footprint	0.14 GB
Max RSS	3.42 GB

Output-identical to a stored reference baseline. All ten guided outputs diff clean against it — every character of every field, in order. That equality is behavioural and exact; it does not mean byte-identical weights, since .aimodel conversion is not byte-deterministic.

Long context

Needle-in-haystack: 3/3 verbatim at both 8k and 15k. Three distinctive facts planted at 10% / 50% / 90% of the filler, strict scoring (a fact counts only if the distinctive entity comes back correct). All three returned at 8,099 tokens and all three at 15,106 tokens, verbatim, including the date.

probe	prompt tokens	TTFT	decode	wall	peak footprint
needle 8k	8,099	6.98 s	64.5 tok/s	8.3 s	0.21 GB
needle 15k	15,106	12.79 s	59.7 tok/s	14.4 s	0.32 GB

Free-form generation from a fixed prompt at three depths, 900-token cap (the model self-stopped inside it every time):

depth	prompt tokens	TTFT	decode	generated	wall	peak footprint
3.4k	3,371	3.22 s	71.0 tok/s	361	8.5 s	0.16 GB
8k	7,953	6.69 s	63.1 tok/s	683	17.7 s	0.30 GB
12k	12,099	9.97 s	59.4 tok/s	514	19.1 s	0.29 GB

Decode barely moves with depth: 71.0 → 63.1 → 59.4 tok/s across a 3.6× context increase. High context costs prefill time and almost nothing else. TTFT scales cleanly and peak in-process footprint stays under a third of a gigabyte at every depth measured.

Status

Artifact	Status
gpu-pipelined/…_mf64 (ctx 4096)	SHIP — measured, and output-identical to the stored reference baseline.
ctx8192/…_mf64	QUALIFIED AT DEPTH — same weights and graph; measured: 3/3 verbatim needle recall at 8,099 tokens, 63.1 tok/s at 8k, 0.30 GB peak footprint.
ctx16384/…_mf64	QUALIFIED AT DEPTH — 3/3 verbatim needle recall at 15,106 tokens, 59.7 tok/s, 0.32 GB peak footprint, 3.46 GB max RSS.

No PyTorch-reference oracle or PSNR gate has been run. Qualification is behavioural: an exact diff against a stored baseline, plus the depth probes above.

License

LiquidAI/LFM2.5-1.2B-Instruct is released under the LFM Open License v1.0 (lfm1.0). These bundles are a derivative and the same licence travels with them — see the upstream licence. Redistribute the licence with the files and comply with its terms. The contribution here is the conversion and the qualification evidence, not the weights.