BioBasicsBench: GPT-6 Astra vs GPT-6.1 Sol, Astra Wins

OpenAI released BioBasicsBench, a 100-task benchmark built with enzyme-design company Ligo that tests whether AI agents can interpret raw wet-lab data, catch errors, and recognize when evidence cannot support a conclusion. In OpenAI's testing of seven models across 37 settings and 18,500 total results, GPT-6 Astra led at 66.7% mean reward at the highest reasoning effort, ahead of GPT-6.1 Sol at 62.0%, though the 4.7-point gap falls within overlapping 95% confidence intervals. OpenAI cautions that missing roughly a third of available rubric credit means the scores are not a basis for unsupervised lab judgment.

GPT-6 Astra leads BioBasicsBench with a 66.7% mean reward at the highest reasoning setting, ahead of GPT-6.1 Sol at 62.0% and well ahead of every older OpenAI model tested, according to OpenAI's November 6, 2026 write-up https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ . Astra is the pick if you want the strongest result, but the lead over Sol is narrow and the confidence intervals overlap. The same score also means about a third of the available credit on routine wet-lab tasks was missed, so it is not a basis for unsupervised lab judgment, which is OpenAI's own caution. BioBasicsBench is a benchmark of 100 tasks that test whether an AI agent can interpret raw experimental data, spot errors, and recognise when the evidence cannot support a conclusion OpenAI https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ . It was developed with Ligo, a company focused on enzyme design. The tasks fall into five areas: protein methods, imaging and cytometry, quantification and assays, nucleic acid methods, and molecular cloning. They range from reading a DNA sample's molecular weight off a gel to detailed assay analysis OpenAI https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ . The inputs are deliberately messy. They include gel and blot images, plate photographs, microscopy stacks and time-lapse, sequencing reads, plate-reader exports and recordings of bench work, with reference sheets such as ladder charts and plasmid maps to help interpretation. Gels and blots are the largest category at 31 tasks, followed by protocols and notes at 27, according to the data mix table in the OpenAI post https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ . Categories overlap, so those counts do not sum to 100. Scientists wrote each task from their own experiments and paired it with an expert-written grading rubric OpenAI https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ . Every task then went through three reviews: an internal review by a Ligo scientist who did not write it, an external review by a PhD-level scientist working in a wet lab, and a final internal review. Independent coverage of the release, such as AI Affairs https://www.aiaffairs.com/technology/openai-biobasicsbench-ai-agents-wet-lab-tasks/ , restates the same process. The grading matters for reading the headline number. Mean reward is the average task score as a percentage of the available rubric credit, so partial answers earn partial marks, and the 95% intervals come from bootstrap resampling of whole tasks OpenAI https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ ; Subagentic https://subagentic.ai/posts/biobasicsbench-wet-lab-agents/ . OpenAI ran seven of its models across 37 model and reasoning-effort settings, five attempts per task per setting, for 18,500 results in total. Astra comes first on the xhigh leaderboard, with Sol second and a large drop to the older generations. The scores, intervals and output tokens per task below come from OpenAI's post https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ ; the final column is my own subtraction from those figures. | Model | Mean reward | 95% CI | Output tokens per task | Gap to Astra | |---|---|---|---|---| | GPT-6 Astra | 66.7% | 60.3 to 73.1% | 4,704 | n/a | | GPT-6.1 Sol | 62.0% | 55.6 to 68.4% | 5,238 | 4.7 points | | GPT-5.6 Sol | 43.8% | 37.6 to 50.0% | 9,099 | 22.9 points | | GPT-6 Luna | 38.3% | 32.3 to 44.5% | 13,120 | 28.4 points | | GPT-5.5 | 35.4% | 29.7 to 41.0% | 11,135 | 31.3 points | | GPT-5.4 | 33.2% | 28.3 to 38.4% | 13,717 | 33.5 points | | GPT-5.2 | 31.2% | 26.2 to 36.3% | 27,620 | 35.5 points | The verdict between the top two is Astra, with a caveat. A 4.7-point gap sits inside overlapping intervals, so the data do not separate Astra from Sol with confidence. The real break is between the two GPT-6 models at the top and every older model in the table. For a wider comparison of the two newest model families outside biology, see our GPT-6 Astra vs Claude Fable 5.1 verdict https://dev.to/articles/gpt-6-astra-vs-claude-fable-5-1-verdict-2026 . OpenAI published failure cases alongside the scores, and they are instructive OpenAI https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ : None of these is exotic. They are the mistakes a rushed junior scientist makes, which is exactly why a confident wrong answer is the risk to manage. OpenAI states the point directly: "Higher benchmark scores do not yet mean models can make reliable judgments in the lab." Three examples stand out, all from OpenAI's post https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ . Astra recognised a z-stack, meaning cells imaged at different focal depths, instead of inventing a biological explanation for the differences. It noticed a plate labelled with the wrong antibiotic LB Kan50 and rejected it as a control. And it recognised that a dead-cell dye detector could pick up light from a second dye, so brightness alone could not justify calling cells dead. These are control-checking behaviours rather than pattern reading. The agent is asking whether the evidence is valid before interpreting it. OpenAI's hypothesis, which it presents as an observation and not a proven finding, is that stronger native multimodal capabilities may be key to improving wet-lab analysis. Remember the limits. These are the developer's own results on its own models, with no Anthropic, Google or third-party scores Neuralbrew https://neuralbrew.ai/en/articles/2026/11/openai-biobasicsbench-ai-agents-biology-lab/ . OpenAI also frames tracking these capabilities as relevant to biosafety and biosecurity. Physical automation is moving faster than evidence interpretation. Nature reported on November 4, 2026 that Shanghai Jiao Tong University's AutoDNA self-driving lab operates more than 20 types of instrument across thousands of operations Nature https://www.nature.com/articles/d41586-026-03552-8 . The same week, Google DeepMind added drug-discovery tools to Antigravity MobiHealthNews https://www.mobihealthnews.com/news/google-deepmind-adds-drug-discovery-tools-antigravity . Robots that run experiments need something that reads the results reliably. BioBasicsBench measures that missing layer, routine interpretation of gels, plates and traces, and the answer today is useful but not dependable. For related work on scientific agents, see our comparison of Antigravity cheminformatics against El Agente Farmaco https://dev.to/articles/antigravity-cheminformatics-vs-el-agente-farmaco , the GDPX benchmark results for Opus 5.5 and Gemini 4 Carbon https://dev.to/articles/gdpx-benchmark-opus-5-5-vs-gemini-4-carbon , our look at bacterial versus human particle structure with Claude https://dev.to/articles/vault-particle-structure-bacterial-vs-human-claude , and the broader agentic AI in health care comparison https://dev.to/articles/agentic-ai-in-health-care-compared-2026 . Q: What is BioBasicsBench? A: It is a 100-task benchmark from OpenAI and Ligo, released on November 6, 2026, that tests whether AI agents can interpret raw wet-lab data, find errors and recognise when evidence is insufficient OpenAI https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ . Q: Which model scored highest on BioBasicsBench? A: GPT-6 Astra scored highest at 66.7% mean reward at xhigh effort, with GPT-6.1 Sol second at 62.0% OpenAI https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ . Q: Does 66.7% mean Astra solves two thirds of the tasks? A: No. The metric is mean reward, the average share of available rubric credit, so partial answers count Nerova https://nerova.ai/news/openai-ligo-biobasicsbench-wet-lab-evaluation . Q: Were Claude or Gemini models tested? A: No. Only seven OpenAI models were evaluated, so there are no Anthropic, Google or third-party scores to compare Neuralbrew https://neuralbrew.ai/en/articles/2026/11/openai-biobasicsbench-ai-agents-biology-lab/ . Q: Can I reproduce the results myself? A: Not at present. No public task set or evaluation code has been released Nerova https://nerova.ai/news/openai-ligo-biobasicsbench-wet-lab-evaluation . Q: Are AI agents ready to run lab decisions unsupervised? A: OpenAI says higher benchmark scores do not yet mean models can make reliable judgments in the lab, so human review remains necessary OpenAI https://openai.com/index/measuring-how-ai-agents-handle-wet-lab-science/ . No corrections to date. Figures were checked against OpenAI's primary post on November 9, 2026. This article was produced with AI assistance and reviewed against the primary sources linked above. It contains no medical or patient advice.