Benchmark Almanac

VariantBench

Claude Opus 5 52.8% leads since Jul 2026 Not in model cards

Agents discover and interpret genetic variants from real sequencing analyses. A deterministic grader checks the result.

Part of LatchBio's BioBench suite on benchmarks.bio. The V1 set has 118 evaluations. Release month is the GitHub repo creation month.

Made by LatchBio Released Jul 2026 Genomics and life sciences Public tasks plus a private test set Last model added Oct 2026 Scores refreshed 2 Nov 2026
Level
Expert Specialists with years of training, often over hours of work
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
latchbio/variantbench 5 stars Leaderboard benchmarks.bio
27
Models scored on its own leaderboard

Claude Opus 5 leads with 52.8%

Best score per model from the benchmark's own leaderboard, read on 2 Nov 2026. Where a model ran with several settings or harnesses, we keep its best.

27 models. Scroll sideways for the rest.

  1. 52.8 Claude Opus 5
  2. 51.7 Claude Sonnet 5.5
  3. 50.6 Claude Opus 5.5
  4. 50.3 Gemini 4 Argon
  5. 46.6 GPT-6 Astra
  6. 46.0 GPT-6.1 Sol
  7. 45.5 Grok 4.7
  8. 42.4 Claude Opus 4.8
  9. 40.7 Claude Fable 5.1
  10. 40.4 Grok 4.6
  11. 39.8 Claude Sonnet 5
  12. 38.7 GPT-6 Sol
  13. 35.3 GPT-5.6 Sol
  14. 35.0 Claude Opus 4.7
  15. 34.8 GPT-6.1 Terra
  16. 31.4 Kimi K3
  17. 31.4 Gemini 3.7 Flash
  18. 31.4 Gemini 3.8 Flash
  19. 31.1 Claude Sonnet 4.6
  20. 31.1 DeepSeek V4.1 Flash
  21. 30.5 GPT-5.5
  22. 28.0 Mistral Large 4
  23. 27.7 GPT-6 Luna
  24. 27.1 GPT-5.6 Luna
  25. 26.3 GLM-5.3
  26. 17.8 Nemotron 3 Ultra
  27. 14.4 Nemotron 3.5 Lightning 30B-A3B
  1. 1 Claude Opus 5 52.8%
  2. 2 Claude Sonnet 5.5 51.7%
  3. 3 Claude Opus 5.5 50.6%
  4. 4 Gemini 4 Argon 50.3%
  5. 5 GPT-6 Astra 46.6%
  6. 6 GPT-6.1 Sol 46%
  7. 7 Grok 4.7 45.5%
  8. 8 Claude Opus 4.8 42.4%
  9. 9 Claude Fable 5.1 LSVP 40.7%
  10. 10 Grok 4.6 40.4%
  11. 11 Claude Sonnet 5 39.8%
  12. 12 GPT-6 Sol 38.7%
  13. 13 GPT-5.6 Sol 35.3%
  14. 14 Claude Opus 4.7 35%
  15. 15 GPT-6.1 Terra 34.8%
  16. 16 Kimi K3 31.4%
  17. 17 Gemini 3.7 Flash 31.4%
  18. 18 Gemini 3.8 Flash 31.4%
  19. 19 Claude Sonnet 4.6 31.1%
  20. 20 DeepSeek V4.1 Flash 31.1%
  21. 21 GPT-5.5 30.5%
  22. 22 Mistral Large 4 28%
  23. 23 GPT-6 Luna 27.7%
  24. 24 GPT-5.6 Luna 27.1%
  25. 25 GLM-5.3 26.3%
  26. 26 Nemotron 3 Ultra 17.8%
  27. 27 Nemotron 3.5 Lightning 30B-A3B 14.4%
#ModelScoreSettingDateSource
1 Claude Opus 5 52.8% Claude Code Jul 2026 Official leaderboard
2 Claude Sonnet 5.5 51.7% Pi Sep 2026 Official leaderboard
3 Claude Opus 5.5 50.6% Pi Sep 2026 Official leaderboard
4 Gemini 4 Argon 50.3% Pi Sep 2026 Official leaderboard
5 GPT-6 Astra 46.6% Pi Sep 2026 Official leaderboard
6 GPT-6.1 Sol 46% Pi Sep 2026 Official leaderboard
7 Grok 4.7 45.5% Pi Sep 2026 Official leaderboard
8 Claude Opus 4.8 42.4% Claude Code May 2026 Official leaderboard
9 Claude Fable 5.1 40.7% LSVP, Pi Sep 2026 Official leaderboard
10 Grok 4.6 40.4% Pi Aug 2026 Official leaderboard
11 Claude Sonnet 5 39.8% Pi Jun 2026 Official leaderboard
12 GPT-6 Sol 38.7% Pi Sep 2026 Official leaderboard
13 GPT-5.6 Sol 35.3% Pi Jul 2026 Official leaderboard
14 Claude Opus 4.7 35% Pi Apr 2026 Official leaderboard
15 GPT-6.1 Terra 34.8% Codex Oct 2026 Official leaderboard
16 Kimi K3 31.4% Pi Jul 2026 Official leaderboard
17 Gemini 3.7 Flash 31.4% Pi Aug 2026 Official leaderboard
18 Gemini 3.8 Flash 31.4% Pi Sep 2026 Official leaderboard
19 Claude Sonnet 4.6 31.1% Pi Feb 2026 Official leaderboard
20 DeepSeek V4.1 Flash 31.1% Pi Sep 2026 Official leaderboard
21 GPT-5.5 30.5% Pi Apr 2026 Official leaderboard
22 Mistral Large 4 28% Pi Oct 2026 Official leaderboard
23 GPT-6 Luna 27.7% Pi Sep 2026 Official leaderboard
24 GPT-5.6 Luna 27.1% Codex Jul 2026 Official leaderboard
25 GLM-5.3 26.3% Pi Aug 2026 Official leaderboard
26 Nemotron 3 Ultra 17.8% Pi Jun 2026 Official leaderboard
27 Nemotron 3.5 Lightning 30B-A3B 14.4% Pi Aug 2026 Official leaderboard

2 models have held number one since its release in Jul 2026

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

Anthropic
Own leaderboard 2 leaders
Claude Opus 5 52.8%

Scores over time

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

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Opus 5 Claude Opus 4.8 Claude Opus 4.7 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Opus 5 Claude Opus 4.8

No lab model card reports it yet

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.

More in Biology and chemistry

All 96 in Biology and chemistry
  • SUPERChem In use Models solve 500 expert-written chemistry reasoning problems, offered in multimodal and text-only forms. Expert solution paths also allow a reasoning path score. Tencent Hy4 preview 66.4%
  • GeneBench Pro In use Agents analyze messy simulated genomics and biology datasets to estimate a decision-relevant quantity across 129 problems. Scored automatically against known answers. GPT-6 Astra 37.1%
  • LifeSciBench In use Models complete 750 expert-written life science research tasks, often with attached figures, tables, or sequence files. Graded against expert rubrics as score and pass rate. GPT-6 Astra 60.3%
  • MolecularIQ In use Models answer questions about molecular graphs, such as counting, indexing and constrained generation, and a symbolic checker verifies each answer. Intern-S2-397B 62.4%
  • TOMG-Bench In use Models generate molecules from natural language requests in three tasks: molecule editing, molecule optimization and customized generation. Each output is checked for validity and for meeting the requested properties. Intern-S2-397B 66.8%
  • BABE Fading Models reason from experimental results in peer-reviewed biology papers to draw conclusions, with causal and cross-scale inference. Seed2.0 Lite (0428) 57.9%