Georgi Gerganov
Georgi Gerganov is a software builder whose AI work includes the ggml tensor library, whisper.cpp and llama.cpp. He studied physics at university, earned a master's in medical physics, and moved into the software industry. In 2022, he began work on GGML and created whisper.cpp as his port of OpenAI's Whisper model. In 2023, he created llama.cpp to run LLaMA models and formed ggml.ai; in 2024, he helped integrate a fix that made Falcon 3 models work in llama.cpp. Since February 2026, he works with Hugging Face's GGML team to support the ggml and llama.cpp community.
Timeline
-
September 2022
GGML and whisper.cpp
In September 2022, he starts work on GGML, a C library for tensor algebra. He also creates whisper.cpp as his port of OpenAI's Whisper model.
-
By 2023
Physics and software work
By 2023, he works in software after studying physics at university and earning a master's in medical physics.
-
March 2023
Creating llama.cpp
In March 2023, he creates llama.cpp as an implementation of LLaMA inference code.
-
2023
Founding ggml.ai
In 2023, he forms ggml.ai, a company dedicated to developing the ggml tensor library.
-
2024
Helping Falcon 3 run
In 2024, he helps integrate a fix that makes Falcon 3 series models work in llama.cpp.
-
Since February 2026
Joining Hugging Face
In February 2026, he joins Hugging Face with the GGML team to support the community behind ggml and llama.cpp.
Sources and credits
Photo: We have no published photo of Georgi Gerganov yet. Send us a free-licensed photo.
- Llama.cpp
- whisper.cpp — Whisper speech recognition in C/C++ — heyaican
- Bringing Whisper and LLaMA to the masses with Georgi Gerganov (Changelog Interviews #532)
- GGML founder Georgi Gerganov launches company, secures pre-seed funding for AI-at-the-edge tensor library — One Year Ago in AI
- Welcome to the Falcon 3 Family of Open Models!
- GGML and llama.cpp join HF to ensure the long-term progress of Local AI