Check a model

Name what you are paying for. The plot is the same Pareto field as the frontier page. The answer is a computation on this catalogue, not an opinion. Bookmark the URL.

9 undominated of 134 · Aug 28, 2026

As of Aug 28, 2026, MiniMax M2.5 is dominated for Balanced on LMArena. Hy3 scores 82.2 higher and costs 69% less, with a covering envelope. 9 of 134 rated, priced standard models are undominated.

Current model

MiniMax M2.5

Balanced

3 tokens in per 1 out

Hy31441.2 · $0.144/MMiniMax M2.51359.0 · $0.473/MLMArena EloEffective $/M · Balanced
134 rated, priced standard models. Chartreuse is the frontier. Cobalt is the model you named, when it is not on the staircase.

Hy3 is both better and cheaper than MiniMax M2.5.

MiniMax M2.5

LMArena Elo

1359.0

Effective $/M · Balanced

$0.473/M

Hy3

LMArena Elo

1441.2

Effective $/M · Balanced

$0.144/M

Envelope-safe replacements

Each row scores at least as high, costs no more, and covers this model’s context, output, modalities, tools, and reasoning. A cheaper narrower model is not listed here.

ModelLMArenaEffective $/MYou save
Hy31441.2 +82.2$0.144/M69%
DeepSeek V4 Flash 04231431.6 +72.6$0.101/M79%
GPT-5.6 Luna1428.5 +69.5$0.450/M5%
MiMo-V2.51427.3 +68.3$0.175/M63%
Solar Pro 41376.2 +17.2$0.052/M89%
GPT-5.4 Nano1372.8 +13.8$0.463/M2%

Higher score, lower price, named losses

Not a drop-in. The loss is why these are not a recommendation.

ModelLMArenaEffective $/MYou give up
Gemma 4 31B1441.7$0.153/Mmax output
Gemma 4 26B A4B 1434.6$0.138/Mmax output
DeepSeek V3.21424.6$0.290/Mcontext
DeepSeek V3.2 Exp1424.4$0.305/Mcontext, max output
DeepSeek V3.1 Terminus1419.6$0.453/Mcontext, max output

Open the frontier

This page does not mail you. The URL is the watch: same model, same mix, current catalogue.