GLM-5.2 (max) matches Claude Opus 4.8 on Harvey LAB-AA benchmark

Artificial Analysis

All evaluations

Harvey LAB-AA Benchmark Leaderboard

Artificial Analysis' implementation of Harvey's Legal Agent Benchmark (LAB), testing AI agents on real-world legal work from Harvey's dataset of 120 private tasks spanning 24 legal practice areas. The agent reads case documents in a sandbox and produces legal deliverables (e.g., memos, disclosure schedules, deposition summaries), graded criterion-by-criterion by a single LLM rubric judge.

See example tasks

Harvey LAB (Legal Agent Benchmark) is a long-horizon agentic benchmark developed by Harvey to measure how well AI agents perform real legal work rather than answer isolated legal questions.

Each task gives the agent a partner-style instruction and a set of case documents inside a sandbox. The agent reads the materials, works across them, and produces a legal deliverable.

Deliverables are graded criterion-by-criterion against a task-specific rubric by a single LLM judge, so the scores reflect whether the agent satisfied the substantive requirements of the work, not just surface fluency.

The Harvey LAB-AA implementation is run with our Stirrup agent harness on Harvey's dataset of 120 private tasks spanning 24 legal practice areas, and we report both the all-pass rate (the share of tasks where every criterion passes, with no partial credit) and the criterion pass rate (the share of individual rubric criteria the deliverables satisfy).

All evaluations are conducted independently by Artificial Analysis. More information can be found on our Intelligence Benchmarking Methodology page.

Introducing Harvey's Legal Agent Benchmark

Harvey LAB-AA Methodology

harveyai/harvey-labs

ArtificialAnalysis/Stirrup

Harvey LAB-AA

Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) scores the highest on Harvey LAB-AA with a score of 14.2%, followed by Claude Opus 4.8 (Adaptive Reasoning, Max Effort) with a score of 7.5%, and GLM-5.2 (max) with a score of 7.5%

Score

Harvey LAB-AA: All-pass Rate

Share of tasks where every rubric criterion passes (Harvey's all-pass grading, no partial credit) · Independently benchmarked by Artificial Analysis

Reasoning models are indicated by a lightbulb icon

Cost

Harvey LAB-AA: Cost per Task

Average cost per task (USD), broken down by input, cache hit, cache write, reasoning, and answer tokens

Reasoning models are indicated by a lightbulb icon

Average cost per task in the evaluation. Costs are split by input, cache hit, cache write, reasoning, and answer token pricing where canonical token counts are available.

Token Usage

Harvey LAB-AA: Output Tokens per Task

Output tokens used to run one task, broken down by reasoning and answer tokens

Reasoning models are indicated by a lightbulb icon

The average number of answer and reasoning tokens produced per benchmark task in this evaluation.

Speed

Harvey LAB-AA: Time per Task

Weighted average decode time (minutes) per task; excludes TTFT and overhead time · Lower is better

Reasoning models are indicated by a lightbulb icon

The weighted average time (seconds) per evaluation task. This is calculated by dividing output tokens per task by output speed, weighted by the relative weights of each benchmark in the evaluation.

Turns

Harvey LAB-AA Benchmark Leaderboard: Average Turns per Task

Average number of model turns per Harvey LAB-AA task · Lower is better

Reasoning models are indicated by a lightbulb icon

This chart shows the average number of turns the agent takes per task. It is a rough proxy for how many actions, tool calls, and iteration cycles an agent is using to complete benchmark tasks.

Score vs. Release Date

Harvey LAB-AA: All-pass Rate vs. Release Date

Most attractive region

Example Tasks & Submissions

View task set on GitHub

Browse representative Harvey LAB tasks from the public task set, the reference files each model was given, and the deliverables it produced.

Mergers & Acquisitions

Instructions

Review the attached acquisition data room contracts and internal memo for change of control and assignment provisions, and prepare a comprehensive deal team report.

Output: coc-analysis-report.docx

Deliverables

Expected outputs the model must produce

coc-analysis-report.docxA comprehensive deal team report analyzing change of control and assignment provisions across the target’s material contracts.

Reference files

Provided to the model

Open

Open

Open

Open

Open

Open

Open

Open

Open

Open

Open

Open

Open

Open

Open

Open

Open

Open

Open

Model submissions

Deliverables produced by each model

Claude Fable 5 (with fallback) - coc-analysis-report.docx

Open

Explore Evaluations

Artificial Analysis Intelligence Index

A composite benchmark aggregating nine challenging evaluations to provide a holistic measure of AI capabilities across mathematics, science, coding, and reasoning.

Artificial Analysis Openness Index

A composite measure providing an industry standard to communicate model openness for users and developers.

AA-Briefcase: Agentic Knowledge Work Benchmark

A private evaluation developed by Artificial Analysis for frontier agentic capability in long-horizon knowledge work, testing agents on realistic business workflows that require deliverables such as spreadsheets, presentations, and memos.

GDPval-AA v2 Leaderboard

GDPval-AA v2 is Artificial Analysis' evaluation framework for OpenAI's GDPval dataset. It tests AI models on real-world tasks across 44 occupations and 9 major industries. Models are given shell access and web browsing capabilities in an agentic loop via Stirrup to solve tasks, with Elo ratings derived from blind pairwise comparisons.

APEX-Agents-AA Benchmark Leaderboard

Artificial Analysis' implementation of the APEX-Agents benchmark, testing AI agents on long-horizon, cross-application tasks in professional-services environments with realistic application tooling.

AutomationBench-AA: Agentic SaaS Workflow Benchmark

A benchmark measuring agentic task completion across simulated SaaS application environments, scoring the share of each task's objectives completed without guardrail violations.

Harvey LAB-AA Benchmark Leaderboard

Artificial Analysis' implementation of Harvey's Legal Agent Benchmark (LAB), testing AI agents on real-world legal work from Harvey's dataset of 120 private tasks spanning 24 legal practice areas. The agent reads case documents in a sandbox and produces legal deliverables (e.g., memos, disclosure schedules, deposition summaries), graded criterion-by-criterion by a single LLM rubric judge.

𝜏³-Banking Benchmark Leaderboard

A fintech customer-support benchmark from the 𝜏-Knowledge framework that tests whether agents can navigate a large unstructured knowledge base and execute multi-step tool calls to resolve realistic banking workflows.

Terminal-Bench v2.1 Benchmark Leaderboard

A verified refresh of Terminal-Bench v2.0 — 89 curated tasks across software engineering, system administration, data processing, model training, and security, with environment and instruction fixes so scores reflect agent capability rather than environment gaps.

Artificial Analysis Long Context Reasoning Benchmark Leaderboard

A challenging benchmark measuring language models' ability to extract, reason about, and synthesize information from long-form documents ranging from 10k to 100k tokens (measured using the cl100k_base tokenizer).

AA-Omniscience: Knowledge and Hallucination Benchmark

A benchmark measuring factual recall and hallucination across various economically relevant domains.

SciCode Benchmark Leaderboard

A scientist-curated coding benchmark featuring 288 test set subproblems from 80 laboratory problems across 16 scientific disciplines.

Humanity's Last Exam Benchmark Leaderboard

A frontier-level benchmark with 2,500 expert-vetted questions across mathematics, sciences, and humanities, designed to be the final closed-ended academic evaluation.

CritPt Benchmark Leaderboard

A benchmark designed to test LLMs on research-level physics reasoning tasks, featuring 71 composite research challenges.

GPQA Diamond Benchmark Leaderboard

The most challenging 198 questions from GPQA, where PhD experts achieve 65% accuracy but skilled non-experts only reach 34% despite web access.

ITBench-AA Benchmark Leaderboard

Artificial Analysis' implementation of IBM's ITBench benchmark, testing AI agents on Kubernetes incident root-cause analysis from offline incident snapshots. The agent inspects alerts, events, traces, and topology and identifies the contributing-factor entities (deployments, pods, namespaces, network policies, etc.) responsible for the failure.

MMMU-Pro Benchmark Leaderboard

An enhanced MMMU benchmark that eliminates shortcuts and guessing strategies to more rigorously test multimodal models across 30 academic disciplines.

IFBench Benchmark Leaderboard

A benchmark evaluating precise instruction-following generalization on 58 diverse, verifiable out-of-domain constraints that test models' ability to follow specific output requirements.

Terminal-Bench Hard Benchmark Leaderboard

An agentic benchmark evaluating AI capabilities in terminal environments through software engineering, system administration, and data processing tasks.

𝜏²-Bench Telecom Benchmark Leaderboard

A dual-control conversational AI benchmark simulating technical support scenarios where both agent and user must coordinate actions to resolve telecom service issues.

MMLU-Pro Benchmark Leaderboard

An enhanced version of MMLU with 12,000 graduate-level questions across 14 subject areas, featuring ten answer options and deeper reasoning requirements.

LiveCodeBench Benchmark Leaderboard

A contamination-free coding benchmark that continuously harvests fresh competitive programming problems from LeetCode, AtCoder, and CodeForces, evaluating code generation, self-repair, and execution.

MATH-500 Benchmark Leaderboard

A 500-problem subset from the MATH dataset, featuring competition-level mathematics across six domains including algebra, geometry, and number theory.

AIME 2025 Benchmark Leaderboard

All 30 problems from the 2025 American Invitational Mathematics Examination, testing olympiad-level mathematical reasoning with integer answers from 000-999.

Global-MMLU-Lite Benchmark Leaderboard

A lightweight, multilingual version of MMLU, designed to evaluate knowledge and reasoning skills across a diverse range of languages and cultural contexts.