Which model leads RULER 64K?

Q: Which model leads RULER 64K?

Nemotron 3 Ultra 550B A55B Base tops the board at 95.3

RULER 64K Leaderboard — 13 Models Ranked

Pro Mode — raw scores, one benchmark under the lensthird-party eval harness

The field · 13 measured

16.1295.3

13 models measured, most on the third-party eval harness. Nemotron 3 Ultra 550B A55B Base tops the board at 95.3.

13 measured·Lifecycle Updated Oct 7, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 13Bars show 16.12–95.3 · ◆ solid = first-party or better · outlined = arm's-length

01Nemotron 3 Ultra 550B A55B BaseNVIDIA95.3———

02Claude 3.5 SonnetAnthropic95.2$3.0/M$15.0/M$0.095

03MiniMax Text 01MiniMax94.3———

04Gemini 1.5 ProGoogle93.8———

05Kimi K2 BaseMoonshot93.79———

06Gemini 2.0 Flash ExpGoogle93.7———

07DeepSeek-V3.2-Exp-BaseDeepSeek93.3———

08Mistral Large 3 675B Base 2512Mistral90.11———

09Granite 4.2 30B+1 altIBM89.96VENDOR$0.16/M$0.65/M$0.005

10GPT-4oOpenAI88.4$5.0/M$15.0/M$0.113

11Granite 4.2 8B+2 altsIBM80.99VENDOR$0.06/M$0.25/M$0.002

12Granite 4.2 3B+2 altsIBM67.52VENDOR$0.03/M$0.12/M$0.001

13GLM-4.5-BaseZ.ai16.12———

◆ 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.

Verification: 13 models scored · 0 independently verified · 8 vendor cross-reference · 5 vendor-reported · 0 with source disagreement. How these tiers are assigned