Maxime Labonne is Head of Post-Training at Liquid AI, where Labonne's work is post-training the company's LFM2 and LFM2.5 models. Labonne is listed among the contributors to the LFM2 technical report, whose models ship with open weights and deployment packages for tools such as llama.cpp and ExecuTorch. Labonne also publishes the LLM Course on GitHub, which includes tutorials on quantizing models with GGUF and llama.cpp, GPTQ, and ExLlamaV2, and an AutoQuant notebook for producing quantized versions of a model.123
At a glance
- Current role
- Head of Post-Training
- Company
- Liquid AI
- Sector
- Company
- Work listed
- 2
- Sources cited
- 5
- Profile reviewed
Career
Maxime Labonne is a Google Developer Expert in AI and machine learning. At Liquid AI, Labonne leads post-training for the company's own pre-trained language models, which use a custom architecture.41
Models and research
Labonne is among the contributors to the LFM2 Technical Report, released in November 2025, which describes Liquid AI's LFM2 family of open-weight models built for efficient on-device deployment. Before and alongside that work, Labonne released fine-tuned and merged community models such as AlphaMonarch-7B and NeuralBeagle14-7B, and a collection of abliterated models with an accompanying article on removing refusals.521
Teaching and writing
The LLM Course, published on GitHub, offers roadmaps and Colab notebooks for learning about large language models, along with tool notebooks such as LLM AutoEval, LazyMergekit and LazyAxolotl for automating evaluation, merging and fine-tuning. Labonne also curates LLM Datasets, a list of datasets for fine-tuning.31
Based on the course, Labonne co-wrote LLM Engineer's Handbook, a Packt book covering an LLM application from design to deployment, and is also the author of Hands-On Graph Neural Networks Using Python. Labonne writes technical analyses of LLMs and post-training on a Substack blog.34
Work and contributions
LFM2 and LFM2.5
Post-trains Liquid AI's LFM2 and LFM2.5 models, per Labonne's GitHub profile; listed as a contributor to the LFM2 technical report.12
The LLM Course
Publishes the course, whose quantization section links Labonne's tutorials on GGUF with llama.cpp, 4-bit GPTQ, and ExLlamaV2; its tools section links the AutoQuant notebook.13
Timeline
- Listed among the contributors to the LFM2 Technical Report from Liquid AI.5
Where they work
- Liquid AI 4 people in the wiki
Related people
- Alexander AminiAlso at Liquid AI
- Mathias LechnerAlso at Liquid AI
- Ramin HasaniAlso at Liquid AI
- Dwarkesh PatelTeaching and writing · Language models
- Sasha RushLanguage models · Teaching and writing
- Sebastian RaschkaLanguage models · Teaching and writing
- Zvi MowshowitzTeaching and writing · Language models
- Tatsunori HashimotoLanguage models · Teaching and writing
- Wes RothTeaching and writing · Language models
- AI ExplainedTeaching and writing · Language models
Suggest a correction or removal
Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.
References
5 sources · profile reviewed Oct 21, 2026 · data updated Oct 21, 2026