Closed
Rented by the token
The frontier labs’ best models. You reach them through an app, a subscription or an API. The weights never leave the lab.
Reference · read 14 November 2026
Some models you can only rent. Others you can download and run yourself, if your machine has the memory. Here is what exists, where to get it, and what it takes to run.
Closed
The frontier labs’ best models. You reach them through an app, a subscription or an API. The weights never leave the lab.
Open, large
Weights anyone can download. Running the biggest takes a server; most people use them through a cloud host.
Open, small
A few billion parameters. A laptop runs them offline, privately, for free, and less capably.
These are what coding agents use. Their size is not published. The Frontier snapshot has context windows and knowledge cutoffs for eight labs, dated.
Anthropic
Anthropic's top model; $8 in / $40 out per million tokens. Anthropic recommends Opus 5.5 ($4 / $20) as the default.
Anthropic
Small fast tier; from $0.10 in / $0.50 out per million tokens.
Google's strongest reasoning model; price doubles on 1 Dec 2026 ($2→$4 in, $10→$20 out), see google-gemini-pricing.
(Google)
Faster everyday Gemini 4 model; $1.50 in / $7.50 out per million tokens.
(Google)
OpenAI
OpenAI's named flagship; $10 in / $50 out per million tokens.
(OpenAI)
OpenAI
Cheapest tier; $0.1 in / $0.5 out per million tokens.
(OpenAI)
xAI
xAI's flagship and only listed text model; $2.00 in / $6.00 out per million tokens.
(xAI)
Hugging Face
Fully open including training data recipe. Ollama's smollm page is the older SmolLM, not SmolLM3.
Microsoft
Reasoning-dense small model; 2.5GB on Ollama. Weak on stored facts by its own card.
Alibaba Qwen
Laptop or high-end phone; 3.3-4.0GB on Ollama. 0.8B and 2B sizes also exist.
Google DeepMind
Smallest Gemma 4, for phones; the model Ollama's quickstart pulls first.
Google DeepMind
Built for phones and edge devices; text, image and audio input.
Mistral AI
Edge model; fits 12GB VRAM at FP8. 3B and 14B sizes also exist.
NVIDIA
NVIDIA's retune of Gemma 4 E4B for local voice and device assistants; 7.2-7.9GB on Ollama.
Alibaba Qwen
Newest small Qwen (Qwen3.8 has no small sizes); 6.6-7.6GB on Ollama.
Google DeepMind
Laptop-GPU class; 7.7-8.0GB on Ollama.
OpenAI
Over 15B total but only 3.6B active; runs within 16GB of memory, 14GB on Ollama.
Alibaba Qwen
Single-GPU vision-language model; 18GB download on Ollama. Qwen's 2.4T flagship weights are also open (Qwen3.8-2.4T-A95B).
Google DeepMind
Biggest Gemma 4, aimed at consumer GPUs and workstations; 19-20GB on Ollama. A 26B A4B MoE sibling has 3.8B active.
Meta
Meta's newest open weights are still Llama 4 (Apr 2025). Fits one H100 at int4; Maverick sibling is 400B total.
OpenAI
Fits one 80GB GPU thanks to 4-bit MXFP4 weights; 65GB on Ollama.
NVIDIA
Vision-language model for coding and tools; NVIDIA serves it on 4× H100; 80-90GB on Ollama.
Z.ai (Zhipu)
Multimodal coding and agent model. The bigger GLM-5.3 (753B) uses a custom license. Ollama cloud only.
MiniMax
Multimodal coding and agent model with sparse attention for long context. Ollama cloud only, served at 512K.
DeepSeek
Multimodal MoE built for cheap long agent runs. A 1017 update exists. Ollama cloud only.
Mistral AI
Largest European open-weight model; selling it as a hosted service needs a separate license. Ollama cloud only.
Moonshot AI
Largest open-weight model listed; datacenter only. Ollama offers it as cloud only.
A model has to fit in memory to run at a usable speed. At 4-bit, the common download, each billion parameters takes about 0.57 GB, from llama.cpp’s measured 4.8944 bits per weight (ggml-org). We call a model a fit when its weights take under three quarters of the memory, leaving room for the conversation and the system. Mixture-of-experts models only compute with some of their weights per word, but all of them have to be loaded.
Try any size and precision in Will it fit? on Plate V. For agents, not models, see how much machine you need.
Where the weights live. Each model is a versioned repository of files, with a model card from its maker (Hugging Face). Every “Hugging Face” link above goes to one.
Downloads and runs a model with one command, locally or in its cloud (Ollama). Some of the largest are offered there only in the cloud; the card says so.
ollama run gpt-oss:20bA desktop app for macOS, Windows and Linux that finds, downloads and chats with models from Hugging Face (LM Studio). The easiest start if you do not want a terminal.
Start small. A 4 to 9 billion parameter model on a 16 GB machine answers quickly and teaches you what local models are good at, and where the frontier still wins.
Chat models are trained to refuse some requests. In 2024, researchers found that across 13 open chat models that refusal runs through a single direction inside the model. Erase it and the model stops refusing; add it and the model refuses even harmless requests (arXiv (Arditi, Obeso, Syed, Paleka, Panickssery, Gurnee, Nanda)).
Abliteration makes that permanent. Someone finds the refusal direction, then edits the weights so no layer can write along it, with no retraining (Maxime Labonne (Hugging Face community blog)). The result is posted as a new model, usually with “abliterated” in its name.
It has costs. In Maxime Labonne’s own test it degraded the model’s quality, and extra training won back most of the loss (Maxime Labonne (Hugging Face community blog)). It also removes refusals across the board, the reasonable ones included, and says nothing about whether the answers are true.
An uncensored model is a different route to a similar place: its maker deletes refusals from the training data and fine-tunes again (Eric Hartford).
Why it matters to everyone else: both papers conclude that safety training on open weights is easy to undo (arXiv (Arditi, Obeso, Syed, Paleka, Panickssery, Gurnee, Nanda); Maxime Labonne (Hugging Face community blog)). Once weights are public, the maker’s guardrails are a default, not a lock. We do not list abliterated downloads here.