Artificial Analysis 上线 VIALS-AA：生命科学视觉推理榜单
161 个任务专考读凝胶图和质粒图这些科研视觉推理，Gemini 4 Carbon 险胜 GPT-6，但 Claude 每任务贵 29 倍，选模型前先看这份榜单。
Artificial Analysis 联合 Handshake AI 推出 VIALS-AA 榜单，评测模型在生命科学视觉推理上的能力。VIALS 基准包含 161 个任务，覆盖凝胶印迹、流式细胞图、质粒图谱、系统发生树等 8 个领域，用开源 harness Stirrup 跑 3 次取 pass^3。结果 Gemini 4 Carbon（high）以 64% 居首，GPT-6 Astra（max）63%、Claude Opus 5.5 61%。成本差距明显：Opus 5.5 每任务 1.64 美元，约为 GPT-6.1 Sol（0.056 美元）的 29 倍，原因是平均 27 轮对 8 轮。开源模型方面 Kimi K3.1 Pro 以 45% 领先，MiMo-V2.6-Pro 43%。
We’re launching VIALS-AA, a leaderboard for Handshake AI’s VIALS benchmark for visual reasoning in life sciences and biotech research
Visual artifacts are central to how scientists analyze experiments and make research decisions, but interpreting them correctly remains a challenge for frontier models. VIALS’ 161 tasks span eight domains and need models to reason over artifacts including gel blots, flow cytometry plots, plasmid maps, and phylogenies. Each task requires models to interpret an artifact and draw a specific scientific conclusion.
We evaluate models by running each task three times in Stirrup, our open source agent harness, with code execution and image inspection tools. GPT-6 Luna (medium) grades each answer against the reference answer, and we report the share of tasks judged correct in all three repeats (pass^3).
Key takeaways:
➤ The top three models are separated by 3 points: Gemini 4 Carbon (high) from @GoogleDeepMind leads narrowly with 64%, followed by GPT-6 Astra (max) from @OpenAI at 63% and Claude Opus 5.5 (max with fallback) from @AnthropicAI at 61%
➤ Astra and Sol achieve similar scores to Opus and Sonnet at a much lower Cost per Task: Claude Opus 5.5 (max with fallback) averages $1.64 per task, ~29× as much as Sol (~$0.056). This price difference is significantly driven by turn use - GPT-6 Astra (max) and GPT-6.1 Sol (max) average ~8 turns per task, compared with ~27 for Claude Opus 5.5 (max with fallback) and ~25 for Claude Sonnet 5.5 (max with fallback)
➤ Open weights models trail the highest score by ~19 points: Kimi K3.1 Pro leads open weights models on VIALS-AA at 45%, narrowly ahead of MiMo-V2.6-Pro at 43%
➤ 14 of 161 tasks saw safety refusals on at least one model: Claude Sonnet 5.5 (max with fallback) refused on 6 tasks, the most of any model, and half of the 14 affected tasks saw refusals from multiple models
Thank you to @joinHandshakeAI and the VIALS authors for creating the benchmark and collaborating with us on bringing it to Artificial Analysis!
