Benchmark Almanac

Hard-negative protein binding prediction

GPT-5.6 Sol 7.60% leads since Jun 2026 Only in system cards

Agents with structure files and tools pick the one true binder among five similar designed proteins. Scored as pass@4.

OpenAI Preparedness evaluation on proprietary binding data. The concern threshold is 50 percent.

Made by OpenAI Released Apr 2026 Genomics and life sciences Tasks kept private Last in a system card Sep 2026 Last model added Sep 2026
Level
Research frontier Researchers, on open problems or tasks that take days
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.
4
Models scored in lab cards
6
System cards report it, from 1 lab

GPT-5.6 Sol leads with 7.60% among models released by Sep 2026

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

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

  1. 7.60 GPT-5.6 Sol
  2. 1.48 GPT-5.5
  3. 0.0 GPT-5.5 Pro
  4. 0.0 GPT-6 Astra
  1. 1 GPT-5.6 Sol 7.60%
  2. 2 GPT-5.5 1.48%
  3. 3 GPT-5.5 Pro 0%
  4. 4 GPT-6 Astra 0%
#ModelScoreSettingDateSource
1 GPT-5.6 Sol 7.60% Jun 2026 Model card
2 GPT-5.5 1.48% Apr 2026 Model card
3 GPT-5.5 Pro 0% Apr 2026 Model card
4 GPT-6 Astra 0% Sep 2026 Model card

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

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

OpenAI
Lab claims Not tested since Sep 2026
GPT-5.6 Sol 7.60%

Scores over time

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

Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) GPT-5.6 Sol GPT-5.5 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) GPT-5.6 Sol

Labs report it only in system cards

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

6 system cards report it, from 1 lab

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
3 Sep 2026 GPT-6 Astra system card 0 Hard-negative protein binding prediction, pass@4, retiring evaluation (Table 16) Bio and chem risk
6 Aug 2026 GPT-5.6 Sol system card addendum 5.27 Hard-negative protein binding prediction, pass@4, GPT-5.6 Sol (August) (7.1.1.5) Bio and chem risk
9 Jul 2026 GPT-5.6 Sol system card 7.6 Hard-negative protein binding prediction, pass@4 (Table 2, 9.1.1.6) Bio and chem risk
25 Jun 2026 GPT-5.6 Sol system card 7.6 Hard-negative protein binding prediction, pass@4 (Table 2, 9.1.1.6) Bio and chem risk
23 Apr 2026 GPT-5.5 system card 1.48 Hard negative protein binding prediction, pass@4, corrected August 19, 2026 (Table 12) Bio and chem risk
23 Apr 2026 GPT-5.5 Pro system card 0 Hard negative protein binding prediction, pass@4, GPT-5.5 Pro (Table 12) Bio and chem risk

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%