LocalScore - Local AI Benchmark
localscore.ai · 948 words · saved by 1 readers
LocalScore is an open benchmark which helps you understand how well your computer can handle local AI tasks.
LocalScore - Local AI Benchmark LocalScore is an open benchmark which helps you understand how well your computer can handle local AI tasks. LEADERBOARD Tiny - 1B Small - 8B Medium - 14B Llama 3.2 1B Instruct Q4_K - Medium GPU CPU+GPU CPU ACCELERATOR PROMPT GENERATION TTFT LOCALSCORE NVIDIA RTX 6000 Ada Generation GPU / 47 GB PROMPT 27281 tokens/s GENERATION 411 tokens/s TTFT 53 ms LOCALSCORE 5948 NVIDIA GeForce RTX 4090 GPU / 23 GB PROMPT 25538 tokens/s GENERATION 382 tokens/s TTFT 57 ms LOCALSCORE 5562 NVIDIA GeForce RTX 4090 D GPU / 47 GB PROMPT 24549 tokens/s GENERATION 359 tokens/s TTFT 5
saved by
related reading
- Do Qwen3.6 27B quantizations break the pelican?quesma.com
- KernelBench: Can LLMs Write GPU Kernels?scalingintelligence.stanford.edu
- The Best GPUs for Deep Learning in 2023 — An In-depth Analysistimdettmers.com
- LLM Inference Performance Engineering: Best Practices | Databricks Blogdatabricks.com
- AI Model & API Providers Analysis | Artificial Analysisartificialanalysis.ai
- Localmaxxing - Local LLM Inference Speed Testslocalmaxxing.com
- Kimi-K2.5 Inference Benchmark - Luminalluminal.com
- How to Deploy Your Modelhtdym.sailresearch.com
- PostTrainBenchposttrainbench.com
- ⭐️ Fast LLM Inference From Scratchandrewkchan.dev
- How to Think About GPUs | How To Scale Your Modeljax-ml.github.io
- Open-Source Agentic Inference Benchmark | InferenceXinferencex.semianalysis.com