Which model leads ArtifactsBench?

Q: Which model leads ArtifactsBench?

Ling-3.0-flash tops the board at 77

ArtifactsBench Leaderboard — Model Scores

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

The field · 11 measured

36.477

11 models measured, most on the third-party eval harness. Ling-3.0-flash tops the board at 77.

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 36.4–77 · ◆ solid = first-party or better · outlined = arm's-length

01Ling-3.0-flashInclusionAI77VENDOR$0.07/M$0.22/M$0.002

02GPT-5OpenAI73$1.3/M$10.0/M$0.077

03MiniMax M2MiniMax66.8$0.30/M$1.2/M$0.011

04Claude Sonnet 4.5Anthropic61.5$3.0/M$15.0/M$0.146

05GLM-4.6Z.ai59.8$0.57/M$2.2/M$0.023

06Gemini 2.5 ProGoogle57.7$1.3/M$10.0/M$0.097

07Claude Sonnet 4Anthropic57.3$3.0/M$15.0/M$0.157

08DeepSeek-V3.2DeepSeek55.8$0.28/M$0.42/M$0.006

09Kimi K2 InstructMoonshot54.2$0.57/M$2.3/M$0.026

10Seed 2.1 Pro PreviewByteDance51VENDOR———

11MAI-Code-1-FlashMicrosoft36.4VENDOR———

◆ 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 · 7 vendor cross-reference · 4 vendor-reported · 0 with source disagreement. How these tiers are assigned