Which model leads WildBench?

Q: Which model leads WildBench?

Mistral Large 3 tops the board at 68.5

WildBench Leaderboard — 11 Models Ranked

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

The field · 11 measured

42.468.5

11 models measured, most on the official card harness. Mistral Large 3 tops the board at 68.5.

11 measured·Lifecycle Updated Oct 8, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 11Bars show 42.4–68.5 · ◆ solid = first-party or better · outlined = arm's-length

01Mistral Large 3Mistral68.5$0.50/M$1.5/M$0.015

02Ministral 3 14B+3 altsMistral68.5$0.20/M$0.20/M$0.003

03Ministral 3 8B+3 altsMistral66.8$0.15/M$0.15/M$0.002

04Qwen3 VL 8B Instruct+3 altsAlibaba66.33P$0.18/M$0.70/M$0.007

05Qwen3 14B+3 altsAlibaba65.13P$0.35/M$1.4/M$0.013

06Gemma 3 12B+3 altsGoogle63.23P$0.05/M$0.15/M$0.002

07Ministral 3 3B+3 altsMistral56.8$0.10/M$0.10/M$0.002

08Qwen3 VL 4B Instruct+3 altsAlibaba56.83P———

09Gemma 3 4B+3 altsGoogle49.13P$0.05/M$0.10/M$0.002

10Jamba 1.5 LargeAI2148.5$2.0/M$8.0/M$0.103

11Jamba 1.5 MiniAI2142.4$0.20/M$0.40/M$0.007

◆ 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: 11 models scored · 0 independently verified · 5 vendor cross-reference · 6 vendor-reported · 0 with source disagreement. How these tiers are assigned