Multimodal VisionActive
Trust Score:
93
VIALS-AA
Agentic life-sciences benchmark requiring visual interpretation of scientific images.
Launched: —Refresh: static
Status Assessment (active):
Actively evaluated by Artificial Analysis on current frontier models; no saturation evidence recorded.
Performance Timeline
Longitudinal progression of model scores against human baselines.Performance & Historical Trajectory
Empirical score progression across model release dates and evaluation rounds.
Independent Vendor-Reported
| Model | Score | Date | Source Type | Provenance |
|---|---|---|---|---|
| MiMo-V2.6-Pro | 43.5% | 2026-11-15 | independent | Source ↗ |
| Inkling 2 Small | 42.9% | 2026-11-15 | independent | Source ↗ |
| MiniMax M3.1 Flash | 29.8% | 2026-11-15 | independent | Source ↗ |
| Qwen3.8 Max (0902) | 31.7% | 2026-11-15 | independent | Source ↗ |
| Gemini 3.8 Flash | 41% | 2026-11-15 | independent | Source ↗ |
| Kimi K3.1 Pro | 45.3% | 2026-11-15 | independent | Source ↗ |
| DeepSeek V4.1 Flash 1017 | 40.4% | 2026-11-15 | independent | Source ↗ |
| Gemini 4 Argon | 59.6% | 2026-11-15 | independent | Source ↗ |
| Muse Wonder | 55.3% | 2026-11-15 | independent | Source ↗ |
| Claude Opus 5.5 | 61.5% | 2026-11-15 | independent | Source ↗ |
| Mistral Large 4 | 20.5% | 2026-11-15 | independent | Source ↗ |
| Claude Sonnet 5.5 | 59.6% | 2026-11-15 | independent | Source ↗ |
| Grok 4.7 | 37.9% | 2026-11-15 | independent | Source ↗ |
| Claude Haiku 5.5 | 49.1% | 2026-11-15 | independent | Source ↗ |
| GPT-6 Astra | 62.7% | 2026-11-15 | independent | Source ↗ |
| GPT-6 Luna | 27.3% | 2026-11-15 | independent | Source ↗ |
| GLM 5.3 Flash | 39.8% | 2026-11-15 | independent | Source ↗ |
| Gemini 4 Carbon | 64% | 2026-11-15 | independent | Source ↗ |
| GPT-6.1 Terra | 60.2% | 2026-11-15 | independent | Source ↗ |
| Claude Fable 5.2 | 67.1% | 2026-11-15 | independent | Source ↗ |
| GPT-6.1 Sol | 59% | 2026-11-15 | independent | Source ↗ |
Human Baseline & Difficulty Horizon
Calibrated human reference points, specialist benchmarks, and ceiling thresholds.Measured Human Scoren/aNot established
Baseline Protocol & Interpretation
No peer-reviewed human baseline has been published for this benchmark.
Metric & Scoring Methodology
Verification protocols, aggregation formulas, and specialized metric variants.Primary Metric:
score (%)Scoring Engine:llm-judge
Dataset & Compute Cost
Evaluation volume, public availability, API pricing, and local hardware requirements.Total Dataset Size—Annotated evaluation items
Public Test SetPublicOpenly mirrored on repositories
Access GatingOpen AccessUnrestricted download
Evaluation LicenseOpen AccessDataset usage and redistribution terms
Frontier API Compute Cost:
$5 – $20 USD for full benchmark evaluation run on frontier APIs.
Recommended Local GPU Setup:
1x NVIDIA RTX 4090 (24GB) or A100 (40GB/80GB) via vLLM / SGLang
How to Run & Reproduce
Standardized evaluation protocols, CLI commands, and reproducible runner templates.Prompt Regimezero-shot
Reasoning Modedirect
Sampling Temp0
Pass@k Budgetk = 1
Tools & SandboxPure Text
Scoring Verifierllm-judge
Option AEleutherAI LM-Evaluation-Harness (Open-Weight Models)
lm_eval --model hf --model_args pretrained=<model_path> --tasks vials-aa --batch_size autoOption BOpenCompass Evaluation Framework
opencompass --datasets vials-aa --models <model_config>Python APIDeterministic Inference Loop Snippet
# Standard API Evaluation Loop
from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": prompt}],
temperature=0.0,
)Standardized Reporting Requirement:
When publishing results for VIALS-AA, always report the exact prompt template, few-shot exemplar ordering, sampling temperature (temperature=0), maximum reasoning budget tokens, and the precise timestamped model snapshot ID.
Contamination & Memorization Analysis
Audit of pretraining exposure risks, memorization vectors, and refresh policies.Overall Contamination Risk:MEDIUM
Refresh Cadence:
Static fixed snapshot
Test Set Exposure:
Public on web / HuggingFace