AnythingLLM ditches Ollama for Lemonade, shrinking engine from 1.8GB to 137MB
jfowers_amd · reddit · 2026-10-14
Local LLM app AnythingLLM is switching its runtime from Ollama to AMD's Lemonade (built on llama.cpp) and shipping an embedded SDK. The move cuts the engine from 1.8GB to 137MB while supporting CPU, GPU and NPU with any GGUF model — another sign llama.cpp is becoming the common substrate for on-device inference, and AMD is leaning into local AI via Lemonade.
Related event: AnythingLLM Drops Ollama for Lemonade, Shrinking Engine to 137MB(2 posts)→
More from Infra
- Musk backs Morgan Stanley call: SpaceX's Distributed Inference Cloud is the overlooked growth story — elonmusk · 2026-10-14
- AMD, Intel, Google, Microsoft, HPE, Red Hat share stage at OCP Global Summit on scaling AI — firstadopter · 2026-10-14
- Musk: Funding AI data centers is easy, building them is brutally hard — even for SpaceX — elonmusk · 2026-10-14
- Microsoft's Maia 300 AI board shown off: 4 XPUs per tray — BenBajarin · 2026-10-14
- Ayar Labs and wiwynn show a working CPO system with GUC test XPUs on the show floor — BenBajarin · 2026-10-14
- Agent token spend hits 1-2x payroll as coding agents rewrite engineering norms — nlarusstone · 2026-10-14