Hackernews posts about LLaMA

LLaMA is an AI model developed by Facebook that generates human-like text responses to user input.

RSS feed for LLaMA

Related:Facebook Meta Apple M1 Mac Mark Zuckerberg Cerebras Apache 2.0 Python 3.11 TinyLlama ChatGPT Stanford Alpaca

31
newLlama.cpp on a jailbroken PS5: run any GGUF model natively on the consolegithub.com
▲ 375Anon84 5d 💬 87
32
newFaster prompt lookup drafting in llama.cppjadidbourbaki.github.io
▲ 89pptadversary 25d 💬 12
33
newFourier Analysis: Drawing Llamas with Circlesadekau.github.io
▲ 79cebert 27d 💬 11
34
newLLaMA Now Goes Faster on CPUsjustine.lol
▲ 6luu 18d 💬 3
35
newNew in Llama.cpp: Decision Modelshuggingface.co
▲ 6cuvinny 19d 💬 1
36
newExtract v2.5: Rebuilding extraction agents on LlamaParsewww.llamaindex.ai
▲ 5cheesyFish 20d discuss
37
newSonicBoom – llama.cpp's own kernels, 14% faster decode via a graph compilergithub.com
▲ 4nalaginrut 8d discuss
38
newLlama.cpp and WebGPU = Client-side LLMs [video]www.youtube.com
▲ 4TriteJin 14d discuss
39
newI got 2.2x more tokens per second from llama.cpp on Intel Arcgrigio.org
▲ 4grigio 23d discuss
40
newLlama.cpp Under the Hoodwww.cppdepend.com
▲ 4cppfirst 28d discuss
41
newTransformers now runs llama.cpp quantshuggingface.co
▲ 4mmis1000 28d discuss
42
newJeyzma is TypeSafeAI minus $850M plus Go/TinyGo/llama.cpp in the browserjeyzma.com
▲ 3deadprogram 11d discuss
43
newAI Development on Windows: From PyTorch and Llama.cpp to Windows MLdevblogs.microsoft.com
▲ 3antimora 14d discuss
44
newNew in Llama.cpp: Decision Modelshuggingface.co
▲ 3nreece 18d discuss
45
newShow HN: Jeva.cpp – a llama.cpp fork with JEV-compatible API for all LLMsgithub.com
▲ 3pragmatwice 23d discuss
46
newShow HN: Llama.cpp fork with 2-4x multiGPU speed for MoE models bigger than VRAMgithub.com
▲ 2neuralll 26d 💬 3
47
newPS5LM: Llama.cpp on a jailbroken PS5, run any GGUF model natively on the consolegithub.com
▲ 2zdw 7d discuss
48
newKisoku 1.6B: LLM trained solo from scratch on a TPU grant, matches Llama 3.2 1Bgithub.com
▲ 20ARCH 12d discuss
49
newJava vLLM-like framework claim 90% perfomance of llamacpp inference on Nvidia HWwww.reddit.com
▲ 2mikepapadim 13d discuss
50
newMS announces DeepSeek V4 Flash and Nemotron locally, llama.cpp comes to Win MLpasqualepillitteri.it
▲ 2choult 14d discuss
51
newOpen-Source E2E FHE Implementation for Privacy-Preserving Llama 3 8B Inferencearxiv.org
▲ 2simonpure 24d discuss
52
newCustom Models in Oh My Pi: vLLM, Llama.cpp, SGLang and Moredoug.sh
▲ 2dougcalobrisi 25d discuss
53
newTransformers now runs llama.cpp quantshuggingface.co
▲ 2vertigoruntime 29d discuss
54
newLlama.cpp banned him for accidentally tagging in a fork PRgithub.com
▲ 1martianvoid 24d 💬 2
55
newShow HN: Llamachord – a Suzuki Omnichord simulator in the browserllamachord.com
▲ 1klolol 9d 💬 1
56
newShow HN: A layer on top llama.cpp enabling big MoE streaming on phone and laptopgithub.com
▲ 1HelldezNoBan 28h discuss
57
newIf llama.cpp enables MoE to activate multiple expertswww.openai-hub.com
▲ 1ConteMascetti71 5d discuss
58
newFine-tuned Llama 3.1 8B vs. Claude Sonnet 5 on 437 legal clausesreopenly.com
▲ 1dayoff 8d discuss
59
newAI Development on Windows: From PyTorch and Llama.cpp to Windows MLdevblogs.microsoft.com
▲ 1pjmlp 13d discuss
60
newShow HN: Reflex Engine Beats Both Llama.cpp and vLLM on Cold-Start to TTFTgithub.com
▲ 1leochong 21d discuss

More stories →