Which model leads MMlongBench Doc?

nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 leads the MMlongBench Doc board of 7 measured models on the Vector Wire tracker at 57.5. Every score is source-attributed with a verification tier.

MMlongBench Doc Leaderboard — Model Scores

Refresh Models —Deals —Regulatory —Sources —

Pro Mode — raw scores, one benchmark under the lensofficial card harness

The field · 7 measured

0.56257.5

7 models measured, most on the official card harness. nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 tops the board at 57.5.

7 measured·7 new this week·Lifecycle Updated Sep 2, 2026Basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 7Bars show 0.562–57.5· ◆ solid = first-party or better · outlined = arm's-length

01nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16NEWNVIDIA57.53P$0.07/M$0.30/M$0.003

02Nemotron Nano VL V2NEWNVIDIA383P$0.20/M$0.60/M$0.011

03Qwen3.6 PlusNEWAlibaba0.62$0.50/M$3.0/M$2.823

04Qwen3.5-27BNEWAlibaba0.602$0.30/M$2.4/M$2.243

05Qwen3.5-35B-A3BNEWAlibaba0.595$0.25/M$2.0/M$1.891

06Qwen3.5-122B-A10BNEWAlibaba0.59$0.40/M$3.2/M$3.051

07Qwen3 VL 235B A22B ThinkingNEWAlibaba0.562$0.40/M$4.0/M$3.915

◆ solid = independently verified · aggregator attested · vendor attributed — outlined = cross-reference · source-attributed. Dimmed ◆ = confidence inferred. “+N alts” = alternate disclosures in the drawer with why-demoted. Prices per million tokens; $/point = input price ÷ score. Every row opens its model dossier.