K-Dense BYOK on CompBioBench: Three Models, One Harness
Gemini 3.8 Flash scored 94% and GPT-5.6 SOL 92% on CompBioBench v1, a statistical tie. DeepSeek V4 Pro scored 81% on the same K-Dense BYOK harness (n=100).
16 min read
Topic
Benchmarks
12 K-Dense articles on Benchmarks — research automation, benchmarks, and case studies from the K-Dense team.
Introducing K-Bench 01: Nine Frontier Models, 178 Real Scientific Tasks, and a Lot of Confident Wrong Answers
K-Bench 01 is our internal benchmark built from 178 real scientific tasks. Nine frontier models ran them end to end; only one clears the acceptability bar.
11 min readThe Silent 97%: Introducing the waypoint-bio Agent Skill
A microbiome foundation model silently discards 97% of a malformed table and returns a valid embedding. We measured what an Agent Skill does about that.
21 min readThe Millimetre Problem: Introducing the lab-hardware-cad Agent Skill
Introducing lab-hardware-cad, an open-source Agent Skill that turns any AI agent into a CAD engineer for lab hardware, benchmarked over 98 geometry-scored runs.
17 min readK-Dense Web vs. Claude Science: K-Dense Leads on Execution and Overall Score
A 20-task benchmark found K-Dense Web stronger at producing auditable research bundles, while Claude Science held a narrow scientific-quality lead.
24 min readBenchmarking Google's Omni Flash for Scientific Video
A 35-case K-Dense benchmark of Google's Omni Flash shows stunning scientific video quality, but unreliable text, physics, and factual correctness.
46 min readBenchmarking Nano Banana 2 Lite for Scientific Image Generation
A 240-image benchmark of four scientific image models shows Nano Banana 2 Lite is fastest, while GPT Image 2 still leads on raw figure quality.
45 min readAI Scientists Need Lab Escape Rooms, Not More Exams
The next AI scientist benchmark should be a lab escape room: a fresh hidden world, limited probes, real evidence, and no place for science theater to hide.
13 min readReproduction, Not Generation, Is AI's Killer App for Science
AI is flooding science with claims no one can check. The quieter story: agents now reproduce most published findings — and that is the bigger deal.
8 min readCan an AI Agent Run Your Mass Spec Pipeline? Benchmarking the PyOpenMS Skill
A reproducible 250-run study of feature detection, adduct grouping, quantification, and identification, run by an AI agent with and without the pyOpenMS skill.
14 min readBeyond RDKit: Benchmarking the Rowan Agent Skill Against Experiment
A reproducible study of pKa, logD, tautomers, ADME, and docking run by an AI agent with the Rowan skill, measured against RDKit and experimental ground truth.
9 min readK-Dense Web Scores 90.0% on BixBench-Verified-50
K-Dense Web scored 45/50 on BixBench-Verified-50, a cleaned biology-agent benchmark designed to separate real model mistakes from benchmark noise.
4 min read
