LABBench2: An Improved Benchmark for AI Systems Performing Biology Research (LABBench2)
We show this table for reference; we do not rank on it.
A benchmark of realistic biology-research tasks involving literature, figures, tables, databases, and bioinformatics files.
Benchmark score on LABBench2 — October 17, 2026
We compile the LABBench2 rows from provider self-reports. Gemini 4 Argon leads the table at 88.8%, followed by Gemini 3.8 Flash (86.2%) and Claude Opus 5 (84.2%). We do not use these results to rank models overall.
Gemini 4 Argon
Gemini 3.8 Flash
Claude Opus 5
Anthropic
7 modelsKnowledgeCurrentDisplay onlyUpdated October 17, 2026
Benchmark score table (7 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Gemini 4 ArgonGoogle | 88.8% | Not reported | Closed |
| 2 | Gemini 3.8 FlashGoogle | 86.2% | Not reported | Closed |
| 3 | Claude Opus 5Anthropic | 84.2% | Not reported | Closed |
| 4 | GPT-5.6 SolOpenAI | 82.1% | Not reported | Closed |
| 5 | Gemini 3.7 FlashGoogle | 82.1% | Not reported | Closed |
| 6 | GPT-5.6 TerraOpenAI | 81.2% | Not reported | Closed |
| 7 | Claude Sonnet 5Anthropic | 80.1% | Not reported | Closed |
Among the reported LABBench2 rows, Gemini 4 Argon is first at 88.8%. The third row is 4.6 points behind. The broader top-10 range is 8.7 points, so many of the published results sit in a relatively narrow band.
7 models have been evaluated on LABBench2. The benchmark falls in the Knowledge category. LABBench2 is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About LABBench2
Year
2026
Tasks
Nearly 1,900 biology-research tasks
Format
Aggregate accuracy
Difficulty
Real-world biology research
LABBench2 expands LAB-Bench into nearly 1,900 more realistic scientific-research tasks. Google's Gemini 3.8 evaluation gives models a Linux terminal with bioinformatics tools, Python, R, and internet access, and reports the aggregate accuracy as display-only evidence.
Freshness and provenance
Version
LABBench2 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does LABBench2 measure?
A benchmark of realistic biology-research tasks involving literature, figures, tables, databases, and bioinformatics files.
Which model scores highest on LABBench2?
Gemini 4 Argon by Google currently leads with a score of 88.8% on LABBench2.
How many models are evaluated on LABBench2?
7 AI models have published results on LABBench2 in the BenchLM catalog.
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 5,600+ readers.
One email each week. Unsubscribe anytime.