Knowledge · Benchmark profile

BioMysteryBench Human Difficult

Computational biology challenges with objective answers that remained unsolved by independent human experts.

Reference2026Active5 models

Data verified 1 Aug 2026 · Methodology 2.3.0

Benchmark score on BioMysteryBench Human Difficult

Leaderboard · 5 models

Claude Opus 5 leads at 49.4%, followed by GPT-5.6 Terra (49.4%) and Gemini 3.7 Flash (43.5%).

Display onlyKnowledge2026

#1Closed

Claude Opus 5

Anthropic

claude-opus-5

49.4%

Overall #1 · Context 1M tokens

#2Closed

GPT-5.6 Terra

OpenAI

gpt-5.6-terra

49.4%

Overall #11 · Context 1.05M tokens

#3Closed

Gemini 3.7 Flash

Google

gemini-3.7-flash

43.5%

Overall #10 · Context 1.05M tokens

ViewModelsColours

AnthropicOpenAIGoogle

Lumina

BioMysteryBench Human Difficult leaderboard

Accessible data for BioMysteryBench Human Difficult leaderboard
Model	Rank	Model ID	Provider	Score	Unit
Claude Opus 5	#1	claude-opus-5	Anthropic	49.4	percent
GPT-5.6 Terra	#2	gpt-5-6-terra	OpenAI	49.4	percent
Gemini 3.7 Flash	#3	gemini-3-7-flash	Google	43.5	percent
Gemini 3.6 Flash	#4	gemini-3-6-flash	Google	41.2	percent
Claude Sonnet 5	#5	claude-sonnet-5	Anthropic	34.1	percent

Benchmark score table (5 models)

One best compatible score per canonical product · higher is better

BioMysteryBench Human Difficult leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Claude Opus 5 claude-opus-5	Anthropic	closed	Source carrier	49.4%
#2	GPT-5.6 Terra gpt-5.6-terra	OpenAI	closed	Public reference	49.4%
#3	Gemini 3.7 Flash gemini-3.7-flash	Google	closed	Public reference	43.5%
#4	Gemini 3.6 Flash gemini-3.6-flash	Google	closed	Public reference	41.2%
#5	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	Public reference	34.1%

The top of this snapshot is led by Claude Opus 5 at 49.4%; third place is 5.9 points behind. The top-5 spread is 15.3 points.

Configuration-preserving evidencePublished source rows (6)Exact source labels, configurations, versions and dated observations. Same-snapshot canonical/configuration projections are shown once.

Every distinct dated source-published row for BioMysteryBench Human Difficult, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
GPT-5.6 Terra gpt-5-6-terra-max	GPT-5.6 Terra	49.4%	August 2026 human-difficultBioMysteryBench Linux terminal with bioinformatics tools, Python, R and restricted internet	Public reference · not scored provider-reported	Google DeepMind permanent refresh source ↗Observed 2026-08-13 · checked 2026-08-13
Gemini 3.7 Flash gemini-3-7-flash-medium	Gemini 3.7 Flash	43.5%	August 2026 human-difficultBioMysteryBench Linux terminal with bioinformatics tools, Python, R and restricted internet	Public reference · not scored provider-reported	Google DeepMind permanent refresh source ↗Observed 2026-08-13 · checked 2026-08-13
Gemini 3.6 Flash gemini-3-6-flash-high	Gemini 3.6 Flash	41.2%	August 2026 human-difficultBioMysteryBench Linux terminal with bioinformatics tools, Python, R and restricted internet	Public reference · not scored provider-reported	Google DeepMind permanent refresh source ↗Observed 2026-08-13 · checked 2026-08-13
Claude Sonnet 5 claude-sonnet-5-max	Claude Sonnet 5	34.1%	August 2026 human-difficultBioMysteryBench Linux terminal with bioinformatics tools, Python, R and restricted internet	Public reference · not scored provider-reported	Google DeepMind permanent refresh source ↗Observed 2026-08-13 · checked 2026-08-13
Exact BenchLM registry variant Claude Opus 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 5	49.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 5	49.4%	2026Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27

Compare top models

Claude Opus 5 vs GPT-5.6 Terra GPT-5.6 Terra vs Gemini 3.7 Flash Gemini 3.7 Flash vs Gemini 3.6 Flash Gemini 3.6 Flash vs Claude Sonnet 5

About BioMysteryBench Human Difficult

Definition and scoring

Organisation: Anthropic
Category: Knowledge
Version: 2026
Direction: higher is better
Ranking use: Reference
Contamination risk: Unknown

Computational biology challenges with objective answers that remained unsolved by independent human experts. Every genuine source row stays tied to its exact model label, configuration, benchmark version and evaluation system. The summary chart shows one best compatible score per canonical product; Score 2.0 admits only explicitly mapped, frozen protocols and keeps incompatible configurations separate.

Open benchmark source ↗

FAQ

Quick answers

What does BioMysteryBench Human Difficult measure?

Computational biology challenges with objective answers that remained unsolved by independent human experts.

Which model scores highest on BioMysteryBench Human Difficult?

Claude Opus 5 by Anthropic currently leads with 49.4%.

How many models are evaluated on BioMysteryBench Human Difficult?

5 canonical model products have a qualifying score, backed by 6 distinct dated source rows.

Does this affect overall Lumina rank?

No. This benchmark is display-only and does not enter the overall Lumina composite.

Knowledge · Benchmark profile

BioMysteryBench Human Difficult

Computational biology challenges with objective answers that remained unsolved by independent human experts.

Reference2026Active5 models

Data verified 1 Aug 2026 · Methodology 2.3.0

Benchmark score on BioMysteryBench Human Difficult