Compiled for phones

ExecuTorch 1.4 programs for the XNNPACK CPU backend with a 32,768-token context window. Weights are int4 in groups of 32 with int8 dynamic activations, the layout Arm's KleidiAI kernels accelerate on Arm CPUs with the i8mm and dotprod extensions, which recent flagship and mid-range phones have. No GPU or NPU delegate, so the file is not tied to one chip vendor. They are in the format the OpenWeights Android app runs, and each of the three cards reports the memory and speed measured on a Dimensity 9400.

The Qwen3 file needs about 7 GB for its KV cache at the full window and ran only on a 16 GB phone; the LFM2.5 files need about 1 GB.

Abliterated

Refusal-direction ablation of the LFM2.5 models with heretic: a 200-trial search that trades refusal rate against KL divergence from the original, a chosen point on that Pareto front, weights merged, chat template and tool-calling tokens unchanged. Each repository holds the safetensors weights and, in executorch/, an ExecuTorch export of them made with the same 32k recipe as the compiled models above.

• LFM2.5-1.2B-Instruct-heretic
• LFM2.5-2.6B-heretic

Tool-calling research

From the OpenGrad study of calibrated tool use: QwenGrad-DPO, a Direct Preference Optimization checkpoint of Qwen3.5-2B selected under a pre-registered promotion policy, and its ExecuTorch CPU export, published as exported and pending behavioural evaluation. Research artifacts, not production models.

How the exports are made

executorch-model-exporter is the pipeline being built to export on GitHub-hosted runners with ExecuTorch's own per-family recipes, smoke-test each program with the runner the app uses, and publish here; the models above were exported by hand, with the recipe recorded in each card. The phone measurements are on latency, context window and rerun pages, with raw results.

Contact: [email protected]