Benchmark Almanac

LABBench2 CloningQA

Gemini 3 Pro 42.9% leads since February 2026
Not in model cards

Models answer molecular biology questions using supplied information or retrieval tools.

Edison’s leaderboard compares five distinct models for this task. The suite reports separate results for each input setting.

Released
February 2026
Kind
Research science
Level
Expert
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Run by
Anyone
Based on
LABBench2
Strongest model tested
GPT-5.2 Pro #32 on the capability index
Last model added
February 2026
Scores refreshed
November 21st, 2026
5
Models scored on its own leaderboard

2 models share the top score of 42.9% among models released by February 2026

Best score per model from the benchmark's own leaderboard, read on November 21st, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.

The newest model with a score here came out in February 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Gemini 3 Pro tools,high 42.9%
  2. 2 Claude Opus 4.6 tools,high 42.9%
  3. 3 Claude Opus 4.5 tools,high 28.6%
  4. 4 GPT-5.2 tools,high 28.6%
  5. 5 GPT-5.2 Pro 28.6%

Source: Official LABBench2 CloningQA leaderboard

#ModelScoreSettingReleasedRun onSource
1 Gemini 3 Pro 42.9% tools,high, file Nov 2025 Jan 2026 Official leaderboard
2 Claude Opus 4.6 42.9% tools,high, inject Feb 2026 Mar 2026 Official leaderboard
3 Claude Opus 4.5 28.6% tools,high, inject Nov 2025 Mar 2026 Official leaderboard
4 GPT-5.2 28.6% tools,high, inject Dec 2025 Jan 2026 Official leaderboard
5 GPT-5.2 Pro 28.6% inject Dec 2025 Jan 2026 Official leaderboard

Scores at each effort level

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Gemini 3 Pro file
tools,high
42.9%
Claude Opus 4.6 inject
tools,high
42.9%
Gemini 3 Pro inject
tools,high
35.7%
Claude Opus 4.6 file
tools,high
35.7%
Claude Opus 4.5 inject
tools,high
28.6%
Claude Opus 4.5 retrieve
tools,high
28.6%
GPT-5.2 inject
tools,high
28.6%
GPT-5.2 Pro inject
tools,high
28.6%
Claude Opus 4.6 retrieve
tools,high
28.6%
Gemini 3 Pro retrieve
tools,high
21.4%
GPT-5.2 file
tools,high
21.4%
GPT-5.2 Pro retrieve
tools,high
21.4%

Source: Official LABBench2 CloningQA leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Gemini 3 Pro tools,high file 42.9% 4,200
Claude Opus 4.6 tools,high inject 42.9% 62,900
Gemini 3 Pro file 35.7% 483
Gemini 3 Pro inject 35.7% 381
Gemini 3 Pro tools,high inject 35.7% 7,900
Claude Opus 4.6 tools,high file 35.7% 41,600
Claude Opus 4.5 tools,high inject 28.6% 36,500
Claude Opus 4.5 tools,high retrieve 28.6% 12,300
GPT-5.2 tools,high inject 28.6% 21,500
GPT-5.2 Pro inject 28.6% 12,700
Claude Opus 4.6 tools,high retrieve 28.6% 50,600
Gemini 3 Pro retrieve 21.4% 440
Claude Opus 4.5 retrieve 21.4% 902
GPT-5.2 file 21.4% 612
GPT-5.2 inject 21.4% 702
GPT-5.2 tools,high retrieve 21.4% 12,700
GPT-5.2 Pro tools,high inject 21.4% 4,800
GPT-5.2 Pro retrieve 21.4% 7,100
Claude Opus 4.6 inject 21.4% 5,500
Claude Opus 4.6 retrieve 21.4% 958
Claude Opus 4.5 file 14.3% 730
Claude Opus 4.5 tools,high file 14.3% 22,500
Claude Opus 4.5 inject 14.3% 931
GPT-5.2 tools,high file 14.3% 21,000
GPT-5.2 Pro tools,high retrieve 14.3% 6,500
Gemini 3 Pro tools,high retrieve 7.10% 1,000
Claude Opus 4.6 file 7.10% 1,300

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Google DeepMind
Own leaderboard Not tested since Feb 2026
Gemini 3 Pro 42.9%

Source: Official LABBench2 CloningQA leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3 Pro 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3 Pro

Source: Official LABBench2 CloningQA leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Science and reasoning

All 203 in Science and reasoning
  • GPQA Diamond In use 198 graduate-level 'Google-proof' multiple-choice questions in biology, chemistry and physics. Gemini 4 Carbon 97%
  • Terminal-Bench-Science In use 70 scientific research-workflow tasks across five science areas, run in Harbor containers. Hidden tests give the resolution rate. GPT-6 Astra 68.1%
  • BioMysteryBench In use Bioinformatics problems on anonymized real datasets that require real analysis with tools. Rubric-graded, with human-solvable and human-difficult splits. Claude Fable 5.2 83.7%
  • FrontierScience In use Models solve physics, chemistry, and biology problems in an Olympiad track graded by short answer and a Research track graded by 10-point rubrics. Intern-S2-397B 87%
  • HiPhO In use Models solve theory problems from 2024 and 2025 physics olympiad exams, some with diagrams. Official marking schemes grade answers and steps, mapped to medal levels. Gemini 3 Pro 78.3%
  • LABBench2 In use 11 biology research sub-tasks (literature, figures, tables, patents, trials and more) with tool access. Macro-averaged accuracy. GPT-5.2 60.9%

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 9 months ago, in February 2026, for Claude Opus 4.6. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (5 entries).
  • Pass Models scored 5 models we track.
  • Warning Newest score The newest came 9 months ago, in February 2026, for Claude Opus 4.6.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.2 Pro (#32).
  • Pass Headroom The best is 42.9%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on advances.edisonscientific.com: 27 rows for models we track at the last read.
  • Pass Last read November 21st, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check