Open models are published as files anyone can download: you run them on your own hardware instead of sending your work to someone else’s. Mutinai tracks the models, the machines they fit and the tools that run them, and labels every figure with where it came from.
Download the weights and run them on your own machine — no account, no per-token bill.
What can I run? →A laptop, one graphics card, a Mac or a server. You decide what the model runs on.
Compare hardware →Prompts, code and documents stay on the machine you ran them on.
How local models work →Swap runtimes, shrink a model to fit your memory, or fine-tune one on your own data.
Tools & runtimes →The weights are files you keep. Nobody can withdraw, reprice or quietly change them.
What “open” means →v0.30.0 Highlights This release features 762 commits from 315 contributors (104 new)! New models : DeepSeek-V4.1-Flash ( #56214 , #56228 , #56208 ) with the whole KV stored in MXFP8 through the FlashMLA V4.1 record on SM100 ( #56893 ), DeepGEMM Mega-mHC ( #56962 ), and async…
Community Runs and reviews written by members will appear here, kept separate from the sourced facts everywhere else on this page and never merged into them.
This preview has no sign-in, so nothing here could have been written by a member: the accounts, runs and reviews in its database are fictional sample content, and are left out rather than dressed up as activity. Real runs, reviews and discussion appear here once contributions open.
Open-weight models by what you want to do, and what each one needs to run.
25 models9 families75 downloadsGraphics cards, Apple silicon and complete systems, compared by what they can hold.
14 devices13 reference systemsThe programs that load and serve models: runtimes, interfaces, assistants, fine-tuning.
13 projects6 runtimesHow models are compared, what each test measures, and who reported the score.
11 benchmarks69 resultsPapers and announcements from the labs and projects, on the timeline with everything else.
30 papers & announcementsPlain-language guides: what the words mean, what fits your machine, what to build with.
4 guidesOpen models arrive in families. Start with the family and who builds it, then open a release to see its sizes and variants.
A benchmarkA fixed set of questions every model is given, so their scores can be compared on the same task. is a fixed set of questions every model answers, so scores can be compared. Positions below are relative to the best open result Mutinai holds for that test.
developer-reported what it measures →
developer-reported what it measures →
developer-reported what it measures →
developer-reported what it measures →
estimated smallest download at 8K contextHow much text the model can consider at once, counted in tokens — roughly ¾ of a word each.
1 tracked model accepts images, but Mutinai holds no multimodal benchmark results yet, so there is nothing to rank here. See the models →
Each step is a release that beat the previous best open result. Developer-reported scores.
Memory decides what fits; memory speed decides how fast it answers. Prices are shown only with what they mean and when they were checked.
Mutinai records launch prices (MSRPThe price the manufacturer set at launch. What a shop charges today can be very different.) and, where a retail price has been checked, the date it was checked. It keeps no price history yet, so it shows no price trends.
These chips share one pool of memory with the processor, so what fits is chosen when the machine is bought, not by the chip.
Pick the closest setup — we recommend a download, a runtime, and show memory and speed.
Largest models that fit entirely in its 24 GB GPU, at 8K context.