Instructions to use mlc-ai/LFM2.5-1.2B-Instruct-q4f16_1-MLC with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLC-LLM
How to use mlc-ai/LFM2.5-1.2B-Instruct-q4f16_1-MLC with MLC-LLM:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
LFM2.5-1.2B-Instruct-q4f16_1-MLC
This is Liquid AI's LFM2.5 1.2B Instruct model in MLC format q4f16_1.
It supports text generation in WebLLM.
The package context window is 128,000 tokens, with a prefill chunk size of 1,024. The proposed WebLLM record defaults to 4,096 tokens to limit memory use.
The q4f16_1 variant requires WebGPU's shader-f16 feature.
Matching WebGPU candidates use v0_2_87 and web-runtime 0.28.0-dev2; see library provenance.
Converted to MLC format from the original weights. See LICENSE (LFM Open License v1.0), including the commercial-use limitation in section 5.
Sampling uses the MLC defaults (upstream sets none): temperature 1.0, top_p 1.0, repetition_penalty 1.0. Liquid AI recommends temperature 0.1 and repetition_penalty 1.05.
Conversion and WebGPU sources
- Model:
0f604ada3f76. - TVM release base:
be03dc94b001; used for weight conversion; the local Metal timer patch is recorded in the build information. - Weight-conversion MLC-LLM:
5e824377af7f.
Weights were converted with the sources above; WebGPU libraries were compiled separately with the source revisions in the linked library provenance.
Reproduce the conversion
With the conversion toolchain above installed and the pinned HF revision downloaded:
mlc_llm convert_weight hf/LFM2.5-1.2B-Instruct --quantization q4f16_1 --device cpu -o dist/LFM2.5-1.2B-Instruct-q4f16_1-MLC
mlc_llm gen_config hf/LFM2.5-1.2B-Instruct --quantization q4f16_1 --conv-template lfm2 --context-window-size 128000 --prefill-chunk-size 1024 -o dist/LFM2.5-1.2B-Instruct-q4f16_1-MLC
These are equivalent regeneration commands. The shipped configuration was retained from the original generation and adjusted to the maximum context and 1,024-token chunk; its SHA256 is recorded below in the build information.
mlc-build-info.json records the full revisions and source weight checksum.
See the upstream model card for model capabilities, evaluations, and limitations, and the WebLLM documentation for runtime usage.
- Downloads last month
- 117
Model tree for mlc-ai/LFM2.5-1.2B-Instruct-q4f16_1-MLC
Base model
LiquidAI/LFM2.5-1.2B-Base