Decision models on your GPU
A decision model doesn't write text: it reads the input once and scores the options you give it, like routing a request or picking an action. So its speed is the time per question, not tokens per second, and most fit on a cheap card.
The open models
| Model | Size | Memory | Speed reported | Reported score | License | How to run it |
|---|---|---|---|---|---|---|
| Julia-1Supersonic Labs | 0.144B | 0.7 GB | 3 ms on an RTX PRO 6000 | — | Apache 2.0 | llama.cpp |
| Laya | 0.421B | 1.0 GB | 5 ms on an RTX PRO 6000 | — | Apache 2.0 | llama.cpp |
| Kev-4B | 4B | 3.6 GB | 12 ms on an RTX PRO 6000 | — | Apache 2.0 | llama.cpp |
| lev | 4B | 3.6 GB | 36 ms on an RTX PRO 6000 | — | Apache 2.0 | llama.cpp |
| OpenJev | 27B | 17.4 GB | 43 ms on an RTX PRO 6000 | — | CC BY-NC 4.0: not for commercial use | llama.cpp |
| ClefCloudflare | 27B | 17.4 GB | 209 ms on an H200 (Cloudflare's own test, as a service); estimated here from OpenJev, the same size | — | Apache 2.0 | llama.cpp, Workers AI |
| Perplexity Decider v1.2Perplexity | 27B | 17.4 GB | Not published; estimated here from OpenJev, the same size | 69.42, Jev Decision Index | Apache 2.0 (reported; one source says the repo is private) | Its own code |
| mammoth-1Base Labs | 27B | 18.6 GB | Not published; estimated here from OpenJev, the same size | 71.24 to 75.34, Jev Decision Index (sources differ) | MIT | The open weights |
Memory: at Q4 with an 8K context; estimated from the model's size, which matches the files published. Speed: the median per question as its makers report it, on inputs that differ, so a number from one test isn't comparable with another. Scores: as reported; benchmarks' versions differ. Check a model's license before using it in a product. Details checked November 2026.
The cheapest card that holds each one
Live on eBay| Model | Card | Price now | Per question (est.) |
|---|---|---|---|
| Julia-1 | GeForce RTX 2060 Super 6 GB | $89.09 | 17–37 ms |
| Laya | GeForce RTX 2060 Super 6 GB | $89.09 | 29–62 ms |
| Kev-4B | GeForce RTX 2060 Super 6 GB | $89.09 | 70–149 ms |
| lev | GeForce RTX 2060 Super 6 GB | $89.09 | 209–448 ms |
| OpenJev | Tesla P40 | $270.00 | 223–478 ms |
| Clef | Tesla P40 | $270.00 | 223–478 ms |
| Perplexity Decider v1.2 | Tesla P40 | $270.00 | 223–478 ms |
| mammoth-1 | Tesla P40 | $270.00 | 223–478 ms |
One PCIe card, from the live listings, shipping included. Per question: scaled from the model's speed on an RTX PRO 6000 by the card's compute and memory bandwidth, within about half either way.
Compare two cards
On any GPU's page, "Compare with" shows both cards side by side, with decision models next to language models.