Benchmark Almanac

micro1

We track 14 benchmarks that micro1 made or helped make.

0 in use 3 with public code 12 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
Realm Financial Reasoning Agents analyze financial documents and spreadsheets. A model grades their answers against practitioner rubrics. Finance 56.2% Not in cards
PharmaBench (micro1) Agents write clinical trial documentation inside 8 synthetic drug programs, across 120 tasks written by biostatisticians, medical writers, safety physicians and regulatory staff. Medicine Oct 2026 17.8% Not in cards
RealmPathology Models interpret real anatomic pathology reports and must not claim a diagnosis, stage or treatment the report does not support. Rubrics have 772 positive and 299 penalty criteria. Medicine Oct 2026 78% Not in cards
WellnessBench Models answer young people asking about eating, sexual safety, substances and self harm, in clinician-written single-turn and pushback dialogues. A pass must be both safe and helpful. Wellbeing and sycophancy Oct 2026 66% Not in cards
Dual-Arm Sim Frontier models control a two-armed robot in a simulated operating room from cameras and sensors. Tasks include swapping instruments, replacing a drug cartridge and packing a surgical kit. Simulated robots Oct 2026 59.8 Not in cards
LegalReasonBench Models work long-horizon legal matters built by practicing lawyers, writing memoranda and court filings across dependent assignments, in closed-record and open-research settings. Law Oct 2026 50.8% Not in cards
MissionReport Models write government status reports in Word from working files and a template, across 20 expert-written Department of Defense style assignments. Expert rubrics score each decision. Documents and spreadsheets Oct 2026 64.1% Not in cards
micro1 Medical Index Models are scored on the equal-weight average of micro1 medical benchmarks, currently RealmPathology, PharmaBench and WellnessBench. Medicine Oct 2026 71% Not in cards
RealWorkDecision Agents review a frozen company workspace and predict the decision an employee made. Expert-written rubrics allow valid alternatives. Tasks from real jobs Oct 2026 43.8% Not in cards
Ivo-micro1 ContractBench Models review contracts and return redlines across 350 attorney-written tasks in 35 scenarios and five agreement types. Tasks cover first review and replies to a counterparty. Law Oct 2026 64% Not in cards
RedlineBench Agent acts as counsel at one turn of a SaaS contract negotiation and returns tracked-change redlines. Three LLM judges grade attorney-written rubrics. Law Jun 2026 50.5% Not in cards
LongExtractionBench Extraction systems pull structured values from 225 long table-heavy documents averaging 358 pages and about 88,700 ground-truth fields each. Documents, charts and text in images Jun 2026 70.7% Not in cards
CortexRetrievalBench Retrieval systems rank financial documents for a practitioner’s question. The benchmark checks whether an authoritative primary source appears at the top. Finding facts in long inputs Not in cards
Personal Agent Bench Personal assistants complete everyday tasks using connected accounts. Experts check whether each assistant completes the work and respects the user’s instructions. Documents and spreadsheets Not in cards