Find the best local LLM for your hardware, ranked by benchmarks
Details
- External ID
- 48146369
- Source
- HN
- Company
- —
- Product
- Find the best local LLM for your hardware, ranked by benchmarks
- Website domain
- github.com
- Launched
- May 15, 2026
- Cohort
- —
- Upvotes
- 283
- Upvotes percentile
- 0.975767366720517
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Enrichment
- Theme
- ML inference and model optimization
- Vertical
- Horizontal
- Function
- Analytics & BI
- Audience
- Developer
- AI stance
- Not AI
- Project type
- Commercial product
- Normalized one-liner
- local llm benchmark comparison tool
- Manually corrected
- False
Could you build this?
Yes This is an informational tool and directory that aggregates hardware specifications (VRAM, memory bandwidth) and model metadata (quantization, parameter sizes, benchmark scores) to recommend compatible open-source LLMs. A solo developer can vibe-code this frontend and calculation logic using static JSON data or simple scrapers in a couple of days.
Discussion
20 comments analyzed.
Competitors mentioned: llmfit (GitHub project for model selection), Qwen models (3.6-35B-A3B, 3.5-9B alternatives), Mistral 7B, Gemma 2
Concerns raised: Memory bandwidth estimates vary significantly between browsers (40-80GB/s overshoot), Poor accuracy for MoE models (underestimates feasibility), Definition of 'best' is oversimplified and doesn't account for task-specific strengths, LLM benchmarks lack real-world validity and are easily gamed by developers, Missing some recent models (Qwen3 Coder Next)
Feature requests: Pull sliding_window attention config from HuggingFace model configs, Better support for MoE model detection and performance estimation, Task-specific model recommendations instead of generic 'best' ranking, More comprehensive model coverage including latest releases
Competitors
Other products that read as similar to this one — 96 launches clear the similarity bar, closest 8 shown.
Attention rank: #2 of 97 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 198 days after the earliest competitor.
- Local AI · hn · 2026-02-05 · 5 upvotes · similarity 0.57
- Rapid-MLX · hn · 2026-04-18 · 9 upvotes · similarity 0.54
- PocketOrca-LLM · github · 2026-09-14 · 8 upvotes · similarity 0.53
- local-llms-with-mlx · github · 2026-09-13 · 18 upvotes · similarity 0.50
- BestLLMfor · ph · 2026-09-09 · 2 upvotes · similarity 0.47
- Replacing my OS process scheduler with an LLM · hn · 2025-12-30 · 77 upvotes · similarity 0.47
- We benchmarked 18 LLMs on OCR (7K+ calls) · hn · 2026-04-22 · 5 upvotes · similarity 0.46
- QuickCompare by Trismik · ph · 2026-04-26 · 229 upvotes · similarity 0.44
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a analytics & bi tool for Legal yet.