Which model leads HLE-Verified?

Q: Which model leads HLE-Verified?

Gemini 3.8 Flash tops the board at 54.9

HLE-Verified Leaderboard — Model Scores

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

The field · 8 measured

3154.9

8 models measured, most on the official card harness. Gemini 3.8 Flash tops the board at 54.9.

8 measured·Lifecycle Updated Oct 8, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 8Bars show 31–54.9 · ◆ solid = first-party or better · outlined = arm's-length

01Gemini 3.8 Flash+2 altsGoogle54.9$0.75/M$3.8/M$0.041

02GPT-5.6 SolOpenAI54.5$4.0/M$20.0/M$0.220

03Claude Opus 5Anthropic54.4$5.0/M$25.0/M$0.276

04Gemini 3.7 Flash+2 altsGoogle53.6$0.75/M$3.8/M$0.042

05Gemini 3.6 FlashGoogle51.2$0.75/M$3.8/M$0.044

06GPT-5.6 TerraOpenAI51.1$2.0/M$12.0/M$0.137

07Qwen3 Max (Reasoning)Alibaba37.6$1.2/M$6.0/M$0.096

08Claude Sonnet 5Anthropic31$2.0/M$10.0/M$0.194

◆ 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: 8 models scored · 0 independently verified · 4 vendor cross-reference · 4 vendor-reported · 0 with source disagreement. How these tiers are assigned