Andrei

@abetlen

San Francisco, CA

Pinned
Andrei
@abetlen
Jun 8
new openai api compatible webserver for llama-cpp-python - /v1/responses api (w codex cli support) - continuous batching - structured response parsing (auto tools, reasoning) - multimodal support (images, audio) video: gemma 4 12b running in codex on amd ai max desktop
00:00
4
Andrei
@abetlen
Jun 6
btw you can run Gemma 4 QAT using llama.cpp in a T4 Google Colab notebook (link in reply)
Google for Developers
@googledevs
Jun 5
Gemma 4 quantization-aware training (QAT) models are now available, bringing AI performance directly to edge devices and consumer GPUs. These checkpoints are optimized with quantization-aware training to dramatically reduce memory requirements and unlock high-speed local
2
Andrei
@abetlen
Aug 30, 2025
really happy to announce i'm in sf now fulltime on an o1 visa! thank you @jessemhan @morph_labs and @lisawehden for making this happen also, a big thank you to @natfriedman @PhilCulliton and @julien_c
16
Andrei
@abetlen
Jun 12, 2025
Incredibly proud of this work by the @morph_labs team!
This post is unavailable.
1
Andrei
@abetlen
Apr 14, 2025
it's very important that our generally intelligent universal function approximators use the right json protocols to interact with the outside world
3