Hackernews posts about LlamaF

1. 01
   

   newBenchmarking Local LLM Servers: Llama.cpp, Llamafile, LM Studio, and Ollamablog.mozilla.ai

   ▲ 3angpt17d💬 2savesaved
2. 02
   

   newFaster prompt lookup drafting in llama.cppjadidbourbaki.github.io

   ▲ 89pptadversary8d💬 12savesaved
3. 03
   

   newFourier Analysis: Drawing Llamas with Circlesadekau.github.io

   ▲ 79cebert11d💬 11savesaved
4. 04
   

   newNew in Llama.cpp: Decision Modelshuggingface.co

   ▲ 6cuvinny2d💬 1savesaved
5. 05
   

   newExtract v2.5: Rebuilding extraction agents on LlamaParsewww.llamaindex.ai

   ▲ 5cheesyFish3ddiscusssavesaved
6. 06
   

   newLLaMA Now Goes Faster on CPUsjustine.lol

   ▲ 4luu2d💬 3savesaved
7. 07
   

   newI got 2.2x more tokens per second from llama.cpp on Intel Arcgrigio.org

   ▲ 4grigio7ddiscusssavesaved
8. 08
   

   newLlama.cpp Under the Hoodwww.cppdepend.com

   ▲ 4cppfirst12ddiscusssavesaved
9. 09
   

   newTransformers now runs llama.cpp quantshuggingface.co

   ▲ 4mmis100012ddiscusssavesaved
10. 10
    

    newNew in Llama.cpp: Decision Modelshuggingface.co

    ▲ 3nreece2ddiscusssavesaved
11. 11
    

    newShow HN: Jeva.cpp – a llama.cpp fork with JEV-compatible API for all LLMsgithub.com

    ▲ 3pragmatwice6ddiscusssavesaved
12. 12
    

    newTurn any Llama-server into a jev system one endpointgithub.com

    ▲ 3khimaros14ddiscusssavesaved
13. 13
    

    newSSD-Llama: SSD-Native Inference for Trillion-Parameter Moe on a Consumer PCarxiv.org

    ▲ 3AlmostCosmo7916ddiscusssavesaved
14. 14
    

    newShow HN: Llama.cpp fork with 2-4x multiGPU speed for MoE models bigger than VRAMgithub.com

    ▲ 2neuralll9d💬 3savesaved
15. 15
    

    newAnyone using llama.cpp willing to test LlamaRack?github.com

    ▲ 2Vibecoder_29d💬 3savesaved
16. 16
    

    newOpen-Source E2E FHE Implementation for Privacy-Preserving Llama 3 8B Inferencearxiv.org

    ▲ 2simonpure7ddiscusssavesaved
17. 17
    

    newCustom Models in Oh My Pi: vLLM, Llama.cpp, SGLang and Moredoug.sh

    ▲ 2dougcalobrisi9ddiscusssavesaved
18. 18
    

    newTransformers now runs llama.cpp quantshuggingface.co

    ▲ 2vertigoruntime13ddiscusssavesaved
19. 19
    

    newLlama.cpp banned him for accidentally tagging in a fork PRgithub.com

    ▲ 1martianvoid8d💬 2savesaved
20. 20
    

    newShow HN: Reflex Engine Beats Both Llama.cpp and vLLM on Cold-Start to TTFTgithub.com

    ▲ 1leochong5ddiscusssavesaved
21. 21
    

    newContextMemory – Markdown memory for your llama.cpp/vLLM servergithub.com

    ▲ 1vitorcastro786ddiscusssavesaved
22. 22
    

    newUsing Llama-cpp-Python grammars to generate JSONtil.simonwillison.net

    ▲ 1Bluestein12ddiscusssavesaved
23. 23
    

    newBenchmark Llamacpp on Intel Panther Lake: Vulkan vs. SYCL vs. OpenVINO vs. CPUgrigio.org

    ▲ 1grigio19ddiscusssavesaved
24. 24
    

    newGetting 50 GB/S Back from the Apple Neural Engineeiln.github.io

    ▲ 218eiln25d💬 36savesaved
25. 25
    

    newLaunch HN: Magnitude (YC S25) – Self-optimizing inference engine for agentsgithub.com

    ▲ 194anerli4d💬 99savesaved
26. 26
    

    newShow HN: DOOM in the kernel, or fibers in eBPFayles.github.io

    ▲ 53ayles25d💬 14savesaved
27. 27
    

    newShow HN: I built Otis, a minimal AI agent that runs local models out of the boxtriangllabs.ai

    ▲ 19petrenk0n21d💬 4savesaved
28. 28
    

    newShow HN: Jinfer – AI inference engine for the JVM. AI in a jarqxotic.ai

    ▲ 6mukel20d💬 1savesaved
29. 29
    

    newWhy human syntax breaks LLMs (and how to fix agentic coding)

    ▲ 5aslang27d💬 3savesaved
30. 30
    

    newShow HN: IttyBittyAI – Qwen2.5:0.5B running on a 2017 Asus smartphone

    ▲ 2spottedmarley10ddiscusssavesaved

· edit