53 open-source models selected for local deployment. Each with its use case, the memory it needs and how to install it.
Source: Hugging Face and Ollama. Sizes checked on 6 October 2026.
53 models 8 categories 20 machines compared
Pick a model and a quantisation to see which machines it fits.
Pick a model, a quantisation and a context. The bars show which machines it fits.
Estimated memory needed
Weights = parameters × bits per weight ÷ 8 (Q4_0 4.5 · Q4_K_M 4.85 · Q8_0 8.5 · FP16 16). Cache = MB per token × context tokens. Runtime = 1.5 GB.
Measured with ollama ps on our GB10, real use comes in 1.4 to 1.6 GB below this figure: we would rather you have memory to spare than run short. From each machine we also take off the share the operating system keeps.
Estimated speed: –
From least to most memory. Open a machine for its spec sheet and where to buy it.
Tip:
Get this analysis by email
Analysis saved. We will email it to you.
Next step: Hardware selector ROI calculator Grants
53 models
Click a model for details
8K context (16x MiniLM). Better for long documents. 768 dimensions. Apache 2.0.
Any device
ollama pull nomic-embed-text Top MTEB benchmark scores. 1024 dimensions. MIT license. Best quality per compute for English.
Any device
ollama pull bge-large Great balance of speed and quality. 768 dimensions. MIT license. Good for production RAG.
Any device
ollama pull gte-base 8K context, 1024 dimensions, strong multilingual. Top MTEB scores. Apache 2.0.
Any device
ollama pull snowflake-arctic-embed:l Best open image model. Exceptional text rendering and prompt adherence. NVFP4 variant runs on 8GB GPUs.
RTX 4060 Ti 16GB (FP8) or RTX 4090 (FP16)
Massive LoRA/ControlNet ecosystem. Fast on consumer GPUs. Still the most flexible image model.
RTX 4060 Ti (FP16) or any 8GB+ GPU
Single-step generation: near real-time images. Great for interactive apps and demos.
Any 8GB+ GPU
MMDiT architecture. Better prompt understanding than SDXL. Good text rendering.
RTX 4060 Ti 16GB or Mac Mini M4 24GB
3-stage pipeline is more VRAM efficient than SDXL. Good quality at lower compute cost.
Any 6GB+ GPU
Apache 2.0 licensed FLUX variant. 4-step generation (vs 50 for dev). Commercial use OK.
RTX 4060 Ti 16GB (FP8) or RTX 4090
Fastest open video model. Generates 5-second clips from text or images. NVFP4 optimized. Apache 2.0.
RTX 4060 Ti (FP16) or any 8GB+ GPU
Best open video quality. Longer clips than LTX. Strong motion consistency. Needs more VRAM.
RTX 4090 24GB or A100
No models found
Select 2-3 models for a side-by-side comparison.
For jobs that need a bigger model than your machine holds, this is what the API costs. We only publish prices we have checked on each provider's official page.
| Model | Provider | Input | Output | Context |
|---|---|---|---|---|
| GPT-5 | OpenAI | $1.25 | $10.00 | 400K |
| GPT-5 Mini | OpenAI | $0.25 | $2.00 | 400K |
| Claude Sonnet 5 | Anthropic | $2.00 | $10.00 | 200K |
| Claude Haiku 4.5 | Anthropic | $1.00 | $5.00 | 200K |
Sources: OpenAI · Anthropic. To compare with what it costs locally, use the cost calculator.
Subscribe and we tell you when new guides, templates and workflows go up. One email a week, no spam.
Already published: 79 guides and 25 templates. All free, no signup.
Tell us what you want to run and on what budget. We will tell you which hardware you need, which model fits, and what to expect from it, before you spend anything.
79 free guides · 17 compliance templates
Your question, with your email; we answer it ourselves
Your question and your email:
What brings you here?