Nace AI has released Drex Max, a multimodal decision model for agents that use computers and browsers. The Drex Max multimodal decision model reads text and images and returns calibrated probabilities. It writes no text. Nace team reports 67.59 on the public Decision Index 0.3.2, 6.9 points above TypeSafe’s Jev 1.14. For agent builders, that means one cheap, typed call per click.
TL;DR
- Size: Parameters not disclosed. Up to 262,144 tokens per question on the API, including image tokens.
- Runs on: Hosted API only (
drex-maxon Nace Console). No open weights or quantized builds announced. - Performance: 67.59 on Decision Index 0.3.2 (public), the top score Nace lists for any decision model.
- Best: 98.2 on BFCL tool calling, and 92.2 on JevBench versus 86.7 for Jev 1.14.
- Worst: 19.0 on HLE, and 72.5 on MMLU-Pro versus 82.1 for Jev 1.14.
What is Drex Max?
Drex Max is a decision model from Nace AI that answers typed questions about text and images. It is the most accurate model in the Drex family, and the only one that reads images.
A decision model does not generate tokens. You send a state and named questions. Each question is one of 3 types: noul (yes/no), choice (pick a label) or score (ordered scale). The model returns a probability distribution for each answer in 1 HTTP call. Drex can only answer with the options you gave it.
How does Drex Max work with images?
Drex Max uses the same POST /v1/systemone endpoint as every Drex model. You select it with "model": "drex-max". It is a pinned ID, so drex-latest does not route to it.
Images go in an images array. Every question sees the same images, before the state and questions. You refer to them as ‘Image 1’ and ‘Image 2’ in plain text. Each entry is either an uploaded workspace file or inline base64.
Key image rules from the Drex Max guide:
- Formats: still PNG, JPEG or WebP.
- Up to 8 images per request, 16 MiB each, 32 MiB total.
- Each image is resized to at most 1,638,400 pixels (1,280 × 1,280).
- Each image costs up to 1,602 tokens in every question.
- Up to 16 questions per request, with a 4 MiB request body.
How does Drex Max perform on benchmarks?
All scores below come from Nace’s Drex page. They are self-reported. Nace says Drex trains on the index’s official training splits and is scored on held-out splits.
- Decision benchmarks: Drex Max scores 67.6 on the public Decision Index 0.3.2. Jev 1.14 scores 60.7 and Drex 2.0 scores 61.3. On JevBench, Drex Max reaches 92.2 against 86.7 for Jev 1.14.
- Agentic benchmarks: Drex Max posts 0.780 on OSWorld-Verified and 98.2 on BFCL. It scores 85.2 on When2Call and 84.8 on RAGTruth.
- Vision: On the Decision Index Vision public suite, Drex Max scores 74.5. GPT-6 Luna Decisions scores 63.9. Mind2Web reaches 88.6 and InfographicVQA reaches 98.8.
- Where it trails: Jev 1.14 leads on MMLU-Pro (82.1 vs 72.5), HLE (19.4 vs 19.0) and ARC-Challenge (97.6 vs 97.4). On Hateful Memes, Drex Max trails GPT-6 Luna Decisions, 70.6 vs 71.2.
‘});
$(‘#mtp-bars’).innerHTML=h;$(‘#mtp-bnote’).textContent=d.note;
requestAnimationFrame(function(){requestAnimationFrame(function(){$$(‘#mtp-bars .fl’).forEach(function(f){f.style.width=f.dataset.w+’%’})})});
$$(‘#mtp-seg button’).forEach(function(b){b.classList.toggle(‘on’,b.dataset.s===k)});setTimeout(post,60)}
$$(‘#mtp-seg button’).forEach(function(b){b.onclick=function(){drawBars(b.dataset.s)}});
/* vs */
var V=[[‘Decision Index 0.3.2′,’67.6′,’60.7’,1],[‘JevBench’,’92.2′,’86.7′,1],[‘Decision Bench’,’94.3′,’92.3′,1],[‘Game Bench win %’,’64.0′,’50.0′,1],[‘BFCL’,’98.2′,’96.4′,1],[‘OSWorld-Verified’,’0.780′,’0.734′,1],[‘Screenshot decision’,’174 ms’,’253 ms’,1],[‘MMLU-Pro’,’72.5′,’82.1′,0],[‘HLE’,’19.0′,’19.4′,0],[‘ARC-Challenge’,’97.4′,’97.6′,0]];
$(‘#mtp-vs’).innerHTML=V.map(function(r){return ‘
‘}).join(”);
/* limits */
var L=[[’16’,’questions per request’,’Same limit with or without images.’],[‘262,144′,’tokens per question’,’State plus that question, image tokens included. Nace’s landing page lists 131K context.’],[‘8′,’images per request’,’Still PNG, JPEG or WebP. 16 MiB each, 32 MiB total.’],[‘1,602′,’max tokens per image’,’Charged in every question. Images resize to at most 1,638,400 pixels (1,280 × 1,280).’],[‘4 MiB’,’request body’,’Inline base64 images count toward it, so about 3 MiB of image bytes fit.’],[‘$0′,’for failed requests’,’Only a 200 response is billed. Output tokens are never billed.’]];
$(‘#mtp-lims’).innerHTML=L.map(function(l){return ”}).join(”);
$$(‘.lim’).forEach(function(c){function t(){c.classList.toggle(‘open’);setTimeout(post,380)}c.onclick=t;c.onkeydown=function(e){if(e.key===’Enter’||e.key===’ ‘){e.preventDefault();t()}}});
/* calc */
function calc(){var n=+$(‘#mtp-dec’).value,im=+$(‘#mtp-img’).value,tx=+$(‘#mtp-txt’).value,tok=n*(tx+im*1602),c=tok/1e6*0.06;
$(‘#mtp-decv’).textContent=n.toLocaleString(‘en-US’);$(‘#mtp-imgv’).textContent=im;$(‘#mtp-txtv’).textContent=tx.toLocaleString(‘en-US’);
$(‘#mtp-cost’).textContent=”$”+(c<10?c.toFixed(2):Math.round(c).toLocaleString(‘en-US’));$(‘#mtp-tok’).textContent=tok.toLocaleString(‘en-US’)+’ input tokens per day’}
[‘#mtp-dec’,’#mtp-img’,’#mtp-txt’].forEach(function(s){$(s).oninput=calc});calc();
window.addEventListener(‘load’,post);window.addEventListener(‘resize’,post);setTimeout(post,300);setTimeout(post,1200);
})();
