LAB-Bench (v1)

Sources

1. LAB-Bench: Measuring Capabilities of Language Models for Biology Research — published result table ↗

   Jon M. Laurent, Joseph D. Janizek, Michael Ruzo, Michaela M. Hinks, Michael J. Hammerling, Siddharth Narayanan et al. · 2024

   All 9 authors

   Jon M. Laurent, Joseph D. Janizek, Michael Ruzo, Michaela M. Hinks, Michael J. Hammerling, Siddharth Narayanan, Manvitha Ponnapati, Andrew D. White, Samuel G. Rodriques

   Dates & evaluation details

   LAB-Bench (v1) · Accuracy · LitQA2 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · LitQA2. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SuppQA · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SuppQA. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · FigQA · 2026-10-26

   Published model comparison, 2024. LAB-Bench · FigQA. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · TableQA · 2026-10-26

   Published model comparison, 2024. LAB-Bench · TableQA. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · DbQA_dga_task-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · DbQA_dga_task-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · DbQA_gene_location_task-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · DbQA_gene_location_task-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · DbQA_mirna_targets_task-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · DbQA_mirna_targets_task-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · DbQA_mouse_tumor_gene_sets-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · DbQA_mouse_tumor_gene_sets-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · DbQA_oncogenic_signatures_task-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · DbQA_oncogenic_signatures_task-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · DbQA_tfbs_GTRD_task-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · DbQA_tfbs_GTRD_task-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · DbQA_variant_from_sequence_task-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · DbQA_variant_from_sequence_task-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · DbQA_variant_multi_sequence_task-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · DbQA_variant_multi_sequence_task-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · DbQA_vax_response_task-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · DbQA_vax_response_task-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · DbQA_viral_ppi_task-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · DbQA_viral_ppi_task-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · ProtocolQA · 2026-10-26

   Published model comparison, 2024. LAB-Bench · ProtocolQA. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_ORF-seq-AAid-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_ORF-seq-AAid-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_ORF-seq-AAseq-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_ORF-seq-AAseq-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_ORF-seq-numlen-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_ORF-seq-numlen-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_ORF-transeff-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_ORF-transeff-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_PCR-gene-enzprimers-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_PCR-gene-enzprimers-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_PCR-gene-gibshindprimers-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_PCR-gene-gibshindprimers-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_PCR-gene-gibssmaprimers-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_PCR-gene-gibssmaprimers-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_PCR-geneprimers-enz-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_PCR-geneprimers-enz-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_PCR-len-primers-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_PCR-len-primers-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_PCR-primers-len-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_PCR-primers-len-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_PCR-seq-enzprimers-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_PCR-seq-enzprimers-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_PCR-seq-primers-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_PCR-seq-primers-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_Prop-seq-gcpercent-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_Prop-seq-gcpercent-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_RE-seq-lenfrags-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_RE-seq-lenfrags-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · SeqQA_RE-seq-numfrags-v1 · 2026-10-26

   Published model comparison, 2024. LAB-Bench · SeqQA_RE-seq-numfrags-v1. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

   LAB-Bench (v1) · Accuracy · CloningScenarios · 2026-10-26

   Published model comparison, 2024. LAB-Bench · CloningScenarios. Category accuracy over public and private items, averaged over three runs. Categories remain separate; no overall score is synthesized. Table 2; source precision and any reported uncertainty retained. These are source-reported observations, not evaluations run by RLTasks.

TRACE THE SCORE · Per-model notes use published breakdowns where available.