How to Read Open-Source LLM Leaderboards Without Lying
Four LLM leaderboards plus the model card: what each measures, the methodology traps that shift the numbers, and a recipe for picking a model.
Tag
#model-selection
Best Local Video Generation Models by VRAM Tier (October 2026)
Open video models sized for real GPUs, with the text encoder counted in and the licence checked. Wan, Hunyuan, LTX-Video, CogVideoX, Mochi and Allegro compared.
How to Choose an Open-Weight Model Family (September 2026)
Qwen, Llama, Mistral, Gemma, DeepSeek, Phi - which family to commit to, what each is good at, and the licence traps that send you back to negotiate.
How to Evaluate Local LLMs on Your Own Workload (September 2026)
Why general benchmarks like MMLU and GPQA don't predict your results, the index-rebase trap, and a recipe for picking a local model from your own prompts.
Best Local Reasoning Models by VRAM (October 2026)
Reasoning-capable open-weight models sized by VRAM. DeepSeek-R1, Qwen3 with thinking, QwQ, Phi-4 reasoning, gpt-oss, and the licence traps.
Open Model Licences for Commercial Use (September 2026)
Apache-2.0, MIT, Qwen variants, Llama community terms: what the Hugging Face API actually returns and the per-repo traps that send you back to negotiate.
Best Local Embedding and Reranker Models by VRAM (September 2026)
Local embedding and reranking models for RAG, sized by VRAM. Real GGUF file sizes, licences, and the runtime gap that stops Ollama reranking.
Best Local Image Generation Models by VRAM (2026)
Open image models sized for real GPUs, with the text encoder counted in and the licence checked. FLUX 3, Z-Image, Qwen-Image 2.1 and SD3.5 compared.
Local AI by VRAM: Which Models Fit Your GPU (September 2026)
Find your VRAM tier, then the current open-weight pick for chat, coding, vision, speech, translation or agents. Eighteen guides, one index.
GGUF vs AWQ vs GPTQ vs MLX: Which Quantization to Use
Quantization formats compared with real file sizes and bits per weight. Why Q4 does not halve a model, and why low quants generate faster.