Which model leads VoiceBench?

Q: Which model leads VoiceBench?

Gemini 3.1 Pro tops the board at 94.3

VoiceBench Leaderboard — 8 Models Ranked

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

The field · 7 measured

63.2894.3

7 models measured, most on the third-party eval harness. Gemini 3.1 Pro tops the board at 94.3.

7 measured·2 new this week·Lifecycle Updated Nov 5, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 7Bars show 63.28–94.3 · ◆ solid = first-party or better · outlined = arm's-length

01Gemini 3.1 ProGoogle94.3$2.0/M$12.0/M$0.074

02InklingNEW+2 altsThinking Machines91.4VENDOR$1.0/M$4.0/M$0.028

03Inkling-SmallNEW+1 altThinking Machines90VENDOR$0.30/M$1.2/M$0.008

04Gemma 4 E4BGoogle72.9$0.02/M$0.10/M$0.001

05Qwen2.5 Omni 3B+1 altAlibaba66.37———

06LFM3-Omni-2BLiquid AI64.73———

07Gemma 4 E2BGoogle63.28———

◆ 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: 7 models scored · 0 independently verified · 4 vendor cross-reference · 3 vendor-reported · 0 with source disagreement. How these tiers are assigned