Downloads · 30 days
0
PrismPhi/lfm2.5-350m-q6a-qcs6490-qnn-npu
lfm2.5-350m-q6a-qcs6490-qnn-npu is a machine learning model from PrismPhi. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as other.
Downloads · 30 days
0
Access
Public
Updated Jul 23, 2026
Repo size
817 MB
Likes
0
Public
Click a slice to open those files.
.onnx717 MB · 87%
From the Hugging Face model README
日本語版 -> MODEL_CARD.ja.md
GitHub repository: radxa-dragon-q6a-qcs6490-lfm2.5-350m-qnn-npu
Unofficial model derivative: This model distribution is not an official, endorsed, affiliated, or sponsored distribution of Liquid AI, Qualcomm, Radxa, Microsoft, OpenAI, or Anthropic.
AI-assisted development disclosure: OpenAI Codex and Anthropic Claude Code were used for research, code generation and editing, debugging, experiment organization, and documentation. A human reviewed the outputs and made the Q6A hardware-validation, adoption, and publication decisions.
A QCS6490/Q6A A16W8 QDQ ONNX derivative of Liquid AI's LFM2.5-350M. The model distribution contains raw QDQ and host-side constants only. Users generate EPContext on their own device.
qdq/chunk16_a16w8_qdq.onnxqdq/decode_a16w8_qdq.onnxhost/embedding_int8_rowwise/*host/rope_cache.npztokenizer/*MODEL_LICENSEasset-manifest.jsonQDQ ONNX files reconstruct the source model into a QNN-oriented graph and apply A16W8 QDQ. Embedding is quantized to rowwise symmetric int8. RoPE cache is mechanically transformed from official ONNX initializers into NPZ.
Local experiments, an OpenAI-compatible API, and OpenWebUI on QCS6490/Q6A. General long-form quality, other HTP generations, and ctx4096 are not guaranteed.
Use the tested one-command installer described in the GitHub Quick Start. It downloads all 11 assets from pinned revision 773ff42cc383cb61ecf32eb13d1f828634fbd0e1, verifies SHA-256, generates QNN EPContexts on the target Q6A, and requires finite QNN-only generation/reload. Its deterministic API canary requires normal Tokyo, JSON {"answer":"Tokyo"}, and first token ID 40550.
The tested stack is QCS6490 HTP v68, ONNX Runtime 1.27.0, and onnxruntime-qnn 2.3.0. EPContext is device/runtime-specific and is not distributed. The installer fingerprints QNN EP/HTP/Stub/Skel hashes and provider/session options; any mismatch regenerates the context.
The released runner measures 17.00-17.60 tok/s API decode and 0.31-1.03 s TTFT. See the validated configuration and measurement conditions on GitHub.
LFM Open License v1.0. Free commercial use is limited to annual revenue below $10,000,000. Redistribution requires including the license, marking changes, and retaining attribution.