whereismyai
whichllm
whichllm is a command-line tool that auto-detects system hardware and ranks the best local large language models from HuggingFace.
Whichllm evaluates local system hardware including GPU, CPU, and RAM to determine which large language models can run effectively. It fetches model data directly from the HuggingFace API and combines it with merged real benchmarks such as LiveBench, Artificial Analysis, Aider, and Chatbot Arena ELO. The tool calculates memory requirements, estimated generation speed, and recency-aware scores to recommend top models. Users can run models instantly, generate Python code snippets, and simulate different GPUs before purchasing hardware. It supports GGUF, AWQ, GPTQ, FP16, and BF16 model formats.
whichllm at a glance
Key features
- Auto-detect hardware for NVIDIA, AMD, Intel, Apple Silicon, and CPU
- Smart ranking based on VRAM fit, speed, and benchmark quality
- One-command chat to download and start interactive sessions
- Code snippets that print ready-to-run Python for any model
- Task profiles to filter by general, coding, vision, or math use cases
- GPU simulation to test performance with different hardware
- JSON output for pipeline and script integration
Use it for
- Finding the best local LLM that fits specific system hardware
- Simulating GPU performance before buying new hardware
- Generating python code snippets to run specific models
- Planning hardware upgrades by comparing current and candidate GPUs
Alternatives to whichllm
- AirLLM — Runs massive 70B+ parameter models on small 4GB consumer GPUs.
- llmrun — Check which open models run on your specific GPU or Mac instantly.
- RunThisLLM — Provides exact hardware and VRAM requirements for over 295 local AI models.
- llmfit — Auto-detects hardware to recommend and score compatible open-source LLMs.
- llm-d — Deploys production-grade LLM serving on existing hardware clusters.
- Soup — Fine-tune and align models like Llama-3.1-8B on a 4 GB laptop GPU.
- MLX LM — Runs LLMs and fine-tuning natively on Apple silicon hardware.
- WebLLM — Runs high-performance language models directly inside web browsers.
- vLLM — Maximizes throughput with PagedAttention and continuous batching.
- LMDeploy — Accelerates inference with high-performance engines and quantization tools.
- MTPLX — It runs local models twice as fast via native MTP speculative decoding.
- Strata — Runs large AI models locally on consumer gaming PCs without server reliance.
All whichllm alternatives